korg/linux
0
0
Fork 0
mirror of https://mirrors.bfsu.edu.cn/git/linux.git synced 2024-11-11 04:18:39 +08:00
Code Issues Packages Projects Releases Wiki Activity
master
linux/fs/nfsd/nfs4proc.c

3741 lines
101 KiB
C
Raw Permalink Normal View History

Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/*
* Server-side procedures for NFSv4.
*
* Copyright (c) 2002 The Regents of the University of Michigan.
* All rights reserved.
*
* Kendrick Smith <kmsmith@umich.edu>
* Andy Adamson <andros@umich.edu>
*
* 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.
* 3. Neither the name of the University nor the names of its
* contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED ``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.
*/
nfsd: fix incorrect umasks We're neglecting to clear the umask after it's set, which can cause a later unrelated rpc to (incorrectly) use the same umask if it happens to be processed by the same thread. There's a more subtle problem here too: An NFSv4 compound request is decoded all in one pass before any operations are executed. Currently we're setting current->fs->umask at the time we decode the compound. In theory a single compound could contain multiple creates each setting a umask. In that case we'd end up using whichever umask was passed in the *last* operation as the umask for all the creates, whether that was correct or not. So, we should just be saving the umask at decode time and waiting to set it until we actually process the corresponding operation. In practice it's unlikely any client would do multiple creates in a single compound. And even if it did they'd likely be from the same process (hence carry the same umask). So this is a little academic, but we should get it right anyway. Fixes: 47057abde515 (nfsd: add support for the umask attribute) Cc: stable@vger.kernel.org Reported-by: Lucash Stach <l.stach@pengutronix.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-03-22 05:19:02 +08:00
#include <linux/fs_struct.h>
[PATCH] knfsd: nfs4: hold filp while reading or writing We're trying to read and write from a struct file that we may not hold a reference to any more (since a close could be processed as soon as we drop the state lock). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-06-24 13:03:13 +08:00
#include <linux/file.h>
nfsd: Add DEALLOCATE support DEALLOCATE only returns a status value, meaning we can use the noop() xdr encoder to reply to the client. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-11-08 03:44:27 +08:00
#include <linux/falloc.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
#include <linux/slab.h>
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
#include <linux/kthread.h>
NFSD: Refactor NFSv4 OPEN(CREATE) Copy do_nfsd_create() to nfs4proc.c and remove NFSv3-specific logic. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-03-29 02:47:34 +08:00
#include <linux/namei.h>
NFSD add COPY_NOTIFY operation Introducing the COPY_NOTIFY operation. Create a new unique stateid that will keep track of the copy state and the upcoming READs that will use that stateid. Each associated parent stateid has a list of copy notify stateids. A copy notify structure makes a copy of the parent stateid and a clientid and will use it to look up the parent stateid during the READ request (suggested by Trond Myklebust <trond.myklebust@hammerspace.com>). At nfs4_put_stid() time, we walk the list of the associated copy notify stateids and delete them. Laundromat thread will traverse globally stored copy notify stateid in idr and notice if any haven't been referenced in the lease period, if so, it'll remove them. Return single netaddr to advertise to the copy. Suggested-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Andy Adamson <andros@netapp.com>
2019-10-05 04:34:26 +08:00
#include <linux/sunrpc/addr.h>
NFSv4.2: Fix NFS4ERR_STALE error when doing inter server copy NFS_FS=y as dependency of CONFIG_NFSD_V4_2_INTER_SSC still have build errors and some configs with NFSD=m to get NFS4ERR_STALE error when doing inter server copy. Added ops table in nfs_common for knfsd to access NFS client modules. Fixes: 3ac3711adb88 ("NFSD: Fix NFS server build errors") Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2020-10-19 11:42:49 +08:00
#include <linux/nfs_ssc.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
#include "idmap.h"
nfsd: Move private headers to source directory Lots of include/linux/nfsd/* headers are only used by nfsd module. Move them to the source directory Signed-off-by: Boaz Harrosh <bharrosh@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-12-04 02:30:56 +08:00
#include "cache.h"
#include "xdr4.h"
nfsd: make fs/nfsd/vfs.h for common includes None of this stuff is used outside nfsd, so move it out of the common linux include directory. Actually, probably none of the stuff in include/linux/nfsd/nfsd.h really belongs there, so later we may remove that file entirely. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-11-05 07:12:35 +08:00
#include "vfs.h"
nfsd41: handle current stateid in open and close Signed-off-by: Tigran Mkrtchyan <kofemann@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-02-14 05:55:24 +08:00
#include "current_stateid.h"
nfsd: use service net instead of hard-coded init_net This patch replaces init_net by SVC_NET(), where possible and also passes proper context to nested functions where required. Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-11-14 23:22:07 +08:00
#include "netns.h"
nfsd: use get_acl and ->set_acl Remove the boilerplate code to marshall and unmarhall ACL objects into xattrs and operate on the posix_acl objects directly. Also move all the ACL handling code into nfs?acl.c where it belongs. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-12-20 21:16:55 +08:00
#include "acl.h"
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
#include "pnfs.h"
nfsd: add trace events For now just a few simple events to trace the layout stateid lifetime, but these already were enough to find several bugs in the Linux client layout stateid handling. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-08-17 08:02:22 +08:00
#include "trace.h"
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
NFSD: Fix sparse warning in nfssvc.c fs/nfsd/nfssvc.c:36:6: warning: symbol 'inter_copy_offload_enable' was not declared. Should it be static? The parameter was added by commit ce0887ac96d3 ("NFSD add nfs4 inter ssc to nfsd4_copy"). Relocate it into the source file that uses it, and make it static. This approach is similar to the nfs4_disable_idmapping, cltrack_prog, and cltrack_legacy_disable module parameters. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-12-19 01:28:23 +08:00
static bool inter_copy_offload_enable;
module_param(inter_copy_offload_enable, bool, 0644);
MODULE_PARM_DESC(inter_copy_offload_enable,
"Enable inter server to server copy offload. Default: false");
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
#ifdef CONFIG_NFSD_V4_2_INTER_SSC
static int nfsd4_ssc_umount_timeout = 900000; /* default to 15 mins */
module_param(nfsd4_ssc_umount_timeout, int, 0644);
MODULE_PARM_DESC(nfsd4_ssc_umount_timeout,
"idle msecs before unmount export from source server");
#endif
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
#define NFSDDBG_FACILITY NFSDDBG_PROC
NFSv4: do exact check about attribute specified Server should return NFS4ERR_ATTRNOTSUPP if an attribute specified is not supported in current environment. Operations CREATE, NVERIFY, OPEN, SETATTR and VERIFY should do this check. This bug is found when do newpynfs tests. The names of the tests that failed are following: CR12 NVF7a NVF7b NVF7c NVF7d NVF7f NVF7r NVF7s OPEN15 VF7a VF7b VF7c VF7d VF7f VF7r VF7s Add function do_check_fattr() to do exact check: 1, Check attribute specified is supported by the NFSv4 server or not. 2, Check FATTR4_WORD0_ACL & FATTR4_WORD0_FS_LOCATIONS are supported in current environment or not. 3, Check attribute specified is writable or not. step 1 and 3 are done in function nfsd4_decode_fattr() but removed to this function now. Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-05-16 16:22:31 +08:00
static u32 nfsd_attrmask[] = {
NFSD_WRITEABLE_ATTRS_WORD0,
NFSD_WRITEABLE_ATTRS_WORD1,
NFSD_WRITEABLE_ATTRS_WORD2
};
static u32 nfsd41_ex_attrmask[] = {
NFSD_SUPPATTR_EXCLCREAT_WORD0,
NFSD_SUPPATTR_EXCLCREAT_WORD1,
NFSD_SUPPATTR_EXCLCREAT_WORD2
};
static __be32
check_attr_support(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
u32 *bmval, u32 *writable)
{
struct dentry *dentry = cstate->current_fh.fh_dentry;
nfsd: opt in to labeled nfs per export Currently turning on NFSv4.2 results in 4.2 clients suddenly seeing the individual file labels as they're set on the server. This is not what they've previously seen, and not appropriate in may cases. (In particular, if clients have heterogenous security policies then one client's labels may not even make sense to another.) Labeled NFS should be opted in only in those cases when the administrator knows it makes sense. It's helpful to be able to turn 4.2 on by default, and otherwise the protocol upgrade seems free of regressions. So, default labeled NFS to off and provide an export flag to reenable it. Users wanting labeled NFS support on an export will henceforth need to: - make sure 4.2 support is enabled on client and server (as before), and - upgrade the server nfs-utils to a version supporting the new "security_label" export flag. - set that "security_label" flag on the export. This is commit may be seen as a regression to anyone currently depending on security labels. We believe those cases are currently rare. Reported-by: tibbs@math.uh.edu Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-01-04 01:30:11 +08:00
struct svc_export *exp = cstate->current_fh.fh_export;
NFSv4: do exact check about attribute specified Server should return NFS4ERR_ATTRNOTSUPP if an attribute specified is not supported in current environment. Operations CREATE, NVERIFY, OPEN, SETATTR and VERIFY should do this check. This bug is found when do newpynfs tests. The names of the tests that failed are following: CR12 NVF7a NVF7b NVF7c NVF7d NVF7f NVF7r NVF7s OPEN15 VF7a VF7b VF7c VF7d VF7f VF7r VF7s Add function do_check_fattr() to do exact check: 1, Check attribute specified is supported by the NFSv4 server or not. 2, Check FATTR4_WORD0_ACL & FATTR4_WORD0_FS_LOCATIONS are supported in current environment or not. 3, Check attribute specified is writable or not. step 1 and 3 are done in function nfsd4_decode_fattr() but removed to this function now. Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-05-16 16:22:31 +08:00
nfsd: clean up supported attribute handling Minor cleanup, no change in behavior. Provide helpers for some common attribute bitmap operations. Drop some comments that just echo the code. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-10-19 02:18:40 +08:00
if (!nfsd_attrs_supported(cstate->minorversion, bmval))
NFSv4: do exact check about attribute specified Server should return NFS4ERR_ATTRNOTSUPP if an attribute specified is not supported in current environment. Operations CREATE, NVERIFY, OPEN, SETATTR and VERIFY should do this check. This bug is found when do newpynfs tests. The names of the tests that failed are following: CR12 NVF7a NVF7b NVF7c NVF7d NVF7f NVF7r NVF7s OPEN15 VF7a VF7b VF7c VF7d VF7f VF7r VF7s Add function do_check_fattr() to do exact check: 1, Check attribute specified is supported by the NFSv4 server or not. 2, Check FATTR4_WORD0_ACL & FATTR4_WORD0_FS_LOCATIONS are supported in current environment or not. 3, Check attribute specified is writable or not. step 1 and 3 are done in function nfsd4_decode_fattr() but removed to this function now. Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-05-16 16:22:31 +08:00
return nfserr_attrnotsupp;
nfsd: clean up supported attribute handling Minor cleanup, no change in behavior. Provide helpers for some common attribute bitmap operations. Drop some comments that just echo the code. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-10-19 02:18:40 +08:00
if ((bmval[0] & FATTR4_WORD0_ACL) && !IS_POSIXACL(d_inode(dentry)))
return nfserr_attrnotsupp;
nfsd: opt in to labeled nfs per export Currently turning on NFSv4.2 results in 4.2 clients suddenly seeing the individual file labels as they're set on the server. This is not what they've previously seen, and not appropriate in may cases. (In particular, if clients have heterogenous security policies then one client's labels may not even make sense to another.) Labeled NFS should be opted in only in those cases when the administrator knows it makes sense. It's helpful to be able to turn 4.2 on by default, and otherwise the protocol upgrade seems free of regressions. So, default labeled NFS to off and provide an export flag to reenable it. Users wanting labeled NFS support on an export will henceforth need to: - make sure 4.2 support is enabled on client and server (as before), and - upgrade the server nfs-utils to a version supporting the new "security_label" export flag. - set that "security_label" flag on the export. This is commit may be seen as a regression to anyone currently depending on security labels. We believe those cases are currently rare. Reported-by: tibbs@math.uh.edu Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-01-04 01:30:11 +08:00
if ((bmval[2] & FATTR4_WORD2_SECURITY_LABEL) &&
!(exp->ex_flags & NFSEXP_SECURITY_LABEL))
return nfserr_attrnotsupp;
nfsd: clean up supported attribute handling Minor cleanup, no change in behavior. Provide helpers for some common attribute bitmap operations. Drop some comments that just echo the code. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-10-19 02:18:40 +08:00
if (writable && !bmval_is_subset(bmval, writable))
return nfserr_inval;
nfsd: add support for the umask attribute Clients can set the umask attribute when creating files to cause the server to apply it always except when inheriting permissions from the parent directory. That way, the new files will end up with the same permissions as files created locally. See https://tools.ietf.org/html/draft-ietf-nfsv4-umask-02 for more details. Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-01-13 03:24:14 +08:00
if (writable && (bmval[2] & FATTR4_WORD2_MODE_UMASK) &&
(bmval[1] & FATTR4_WORD1_MODE))
return nfserr_inval;
NFSv4: do exact check about attribute specified Server should return NFS4ERR_ATTRNOTSUPP if an attribute specified is not supported in current environment. Operations CREATE, NVERIFY, OPEN, SETATTR and VERIFY should do this check. This bug is found when do newpynfs tests. The names of the tests that failed are following: CR12 NVF7a NVF7b NVF7c NVF7d NVF7f NVF7r NVF7s OPEN15 VF7a VF7b VF7c VF7d VF7f VF7r VF7s Add function do_check_fattr() to do exact check: 1, Check attribute specified is supported by the NFSv4 server or not. 2, Check FATTR4_WORD0_ACL & FATTR4_WORD0_FS_LOCATIONS are supported in current environment or not. 3, Check attribute specified is writable or not. step 1 and 3 are done in function nfsd4_decode_fattr() but removed to this function now. Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-05-16 16:22:31 +08:00
return nfs_ok;
}
static __be32
nfsd4_check_open_attributes(struct svc_rqst *rqstp,
struct nfsd4_compound_state *cstate, struct nfsd4_open *open)
{
__be32 status = nfs_ok;
if (open->op_create == NFS4_OPEN_CREATE) {
if (open->op_createmode == NFS4_CREATE_UNCHECKED
|| open->op_createmode == NFS4_CREATE_GUARDED)
status = check_attr_support(rqstp, cstate,
open->op_bmval, nfsd_attrmask);
else if (open->op_createmode == NFS4_CREATE_EXCLUSIVE4_1)
status = check_attr_support(rqstp, cstate,
open->op_bmval, nfsd41_ex_attrmask);
}
return status;
}
NFSv4: ACL in operations 'open' and 'create' should be used ACL in operations 'open' and 'create' is decoded but never be used. It should be set as the initial ACL for the object according to RFC3530. If error occurs when setting the ACL, just clear the ACL bit in the returned attr bitmap. Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-07-06 17:24:16 +08:00
static int
is_create_with_attrs(struct nfsd4_open *open)
{
return open->op_create == NFS4_OPEN_CREATE
&& (open->op_createmode == NFS4_CREATE_UNCHECKED
|| open->op_createmode == NFS4_CREATE_GUARDED
|| open->op_createmode == NFS4_CREATE_EXCLUSIVE4_1);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
static inline void
fh_dup2(struct svc_fh *dst, struct svc_fh *src)
{
fh_put(dst);
dget(src->fh_dentry);
if (src->fh_export)
NFSD: Using exp_get for export getting Don't using cache_get besides export.h, using exp_get for export. Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-06-10 22:06:44 +08:00
exp_get(src->fh_export);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*dst = *src;
}
[PATCH] nfsd: nfs4 code returns error values in net-endian Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:28:59 +08:00
static __be32
[PATCH] knfsd: nfsd4: fix owner-override on open If a client creates a file using an open which sets the mode to 000, or if a chmod changes permissions after a file is opened, then situations may arise where an NFS client knows that some IO is permitted (because a process holds the file open), but the NFS server does not (because it doesn't know about the open, and only sees that the IO conflicts with the current mode of the file). As a hack to solve this problem, NFS servers normally allow the owner to override permissions on IO. The client can still enforce correct permissions-checking on open by performing an explicit access check. In NFSv4 the client can rely on the explicit on-the-wire open instead of an access check. Therefore we should not be allowing the owner to override permissions on an over-the-wire open! However, we should still allow the owner to override permissions in the case where the client is claiming an open that it already made either before a reboot, or while it was holding a delegation. Thanks to Jim Rees for reporting the bug. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-17 15:10:13 +08:00
do_open_permission(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open, int accmode)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
if (open->op_truncate &&
!(open->op_share_access & NFS4_SHARE_ACCESS_WRITE))
return nfserr_inval;
nfsd4: permit read opens of executable-only files A client that wants to execute a file must be able to read it. Read opens over nfs are therefore implicitly allowed for executable files even when those files are not readable. NFSv2/v3 get this right by using a passed-in NFSD_MAY_OWNER_OVERRIDE on read requests, but NFSv4 has gotten this wrong ever since dc730e173785e29b297aa605786c94adaffe2544 "nfsd4: fix owner-override on open", when we realized that the file owner shouldn't override permissions on non-reclaim NFSv4 opens. So we can't use NFSD_MAY_OWNER_OVERRIDE to tell nfsd_permission to allow reads of executable files. So, do the same thing we do whenever we encounter another weird NFS permission nit: define yet another NFSD_MAY_* flag. The industry's future standardization on 128-bit processors will be motivated primarily by the need for integers with enough bits for all the NFSD_MAY_* flags. Reported-by: Leonardo Borda <leonardoborda@gmail.com> Cc: stable@kernel.org Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-25 22:48:39 +08:00
accmode |= NFSD_MAY_READ_IF_EXEC;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (open->op_share_access & NFS4_SHARE_ACCESS_READ)
nfsd: rename MAY_ flags Rename nfsd_permission() specific MAY_* flags to NFSD_MAY_* to make it clear, that these are not used outside nfsd, and to avoid name and number space conflicts with the VFS. [comment from hch: rename MAY_READ, MAY_WRITE and MAY_EXEC as well] Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-06-16 19:20:29 +08:00
accmode |= NFSD_MAY_READ;
[PATCH] knfsd: nfsd4: fix open permission checking We weren't actually checking for SHARE_ACCESS_WRITE, with the result that the owner could open a non-writeable file for write! Continue to allow DENY_WRITE only with write access. Thanks to Jim Rees for reporting the bug. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-17 15:10:14 +08:00
if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE)
nfsd: rename MAY_ flags Rename nfsd_permission() specific MAY_* flags to NFSD_MAY_* to make it clear, that these are not used outside nfsd, and to avoid name and number space conflicts with the VFS. [comment from hch: rename MAY_READ, MAY_WRITE and MAY_EXEC as well] Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-06-16 19:20:29 +08:00
accmode |= (NFSD_MAY_WRITE | NFSD_MAY_TRUNC);
nfsd4: fix share mode permissions NFSv4 opens may function as locks denying other NFSv4 users the rights to open a file. We're requiring a user to have write permissions before they can deny write. We're *not* requiring a user to have write permissions to deny read, which is if anything a more drastic denial. What was intended was to require write permissions for DENY_READ. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-12-02 08:42:57 +08:00
if (open->op_share_deny & NFS4_SHARE_DENY_READ)
nfsd: rename MAY_ flags Rename nfsd_permission() specific MAY_* flags to NFSD_MAY_* to make it clear, that these are not used outside nfsd, and to avoid name and number space conflicts with the VFS. [comment from hch: rename MAY_READ, MAY_WRITE and MAY_EXEC as well] Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-06-16 19:20:29 +08:00
accmode |= NFSD_MAY_WRITE;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
NFSD: remove redundant variable status Return value directly from fh_verify() do_open_permission() exp_pseudoroot() instead of getting value from redundant variable status. Reported-by: Zeal Robot <zealci@zte.com.cn> Signed-off-by: Jinpeng Cui <cui.jinpeng2@zte.com.cn> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-08-31 22:20:02 +08:00
return fh_verify(rqstp, current_fh, S_IFREG, accmode);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
nfsd: move error choice for incorrect object types to version-specific code. If an NFS operation expects a particular sort of object (file, dir, link, etc) but gets a file handle for a different sort of object, it must return an error. The actual error varies among NFS versions in non-trivial ways. For v2 and v3 there are ISDIR and NOTDIR errors and, for NFSv4 only, INVAL is suitable. For v4.0 there is also NFS4ERR_SYMLINK which should be used if a SYMLINK was found when not expected. This take precedence over NOTDIR. For v4.1+ there is also NFS4ERR_WRONG_TYPE which should be used in preference to EINVAL when none of the specific error codes apply. When nfsd_mode_check() finds a symlink where it expected a directory it needs to return an error code that can be converted to NOTDIR for v2 or v3 but will be SYMLINK for v4. It must be different from the error code returns when it finds a symlink but expects a regular file - that must be converted to EINVAL or SYMLINK. So we introduce an internal error code nfserr_symlink_not_dir which each version converts as appropriate. nfsd_check_obj_isreg() is similar to nfsd_mode_check() except that it is only used by NFSv4 and only for OPEN. NFSERR_INVAL is never a suitable error if the object is the wrong time. For v4.0 we use nfserr_symlink for non-dirs even if not a symlink. For v4.1 we have nfserr_wrong_type. We handle this difference in-place in nfsd_check_obj_isreg() as there is nothing to be gained by delaying the choice to nfsd4_map_status(). As a result of these changes, nfsd_mode_check() doesn't need an rqstp arg any more. Note that NFSv4 operations are actually performed in the xdr code(!!!) so to the only place that we can map the status code successfully is in nfsd4_encode_operation(). Signed-off-by: NeilBrown <neilb@suse.de> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-08-14 21:21:01 +08:00
static __be32 nfsd_check_obj_isreg(struct svc_fh *fh, u32 minor_version)
nfsd4: return nfserr_symlink on v4 OPEN of non-regular file Without this, an attempt to open a device special file without first stat'ing it will fail. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-16 04:55:02 +08:00
{
VFS: normal filesystems (and lustre): d_inode() annotations that's the bulk of filesystem drivers dealing with inodes of their own Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-03-18 06:25:59 +08:00
umode_t mode = d_inode(fh->fh_dentry)->i_mode;
nfsd4: return nfserr_symlink on v4 OPEN of non-regular file Without this, an attempt to open a device special file without first stat'ing it will fail. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-16 04:55:02 +08:00
if (S_ISREG(mode))
return nfs_ok;
if (S_ISDIR(mode))
return nfserr_isdir;
nfsd: move error choice for incorrect object types to version-specific code. If an NFS operation expects a particular sort of object (file, dir, link, etc) but gets a file handle for a different sort of object, it must return an error. The actual error varies among NFS versions in non-trivial ways. For v2 and v3 there are ISDIR and NOTDIR errors and, for NFSv4 only, INVAL is suitable. For v4.0 there is also NFS4ERR_SYMLINK which should be used if a SYMLINK was found when not expected. This take precedence over NOTDIR. For v4.1+ there is also NFS4ERR_WRONG_TYPE which should be used in preference to EINVAL when none of the specific error codes apply. When nfsd_mode_check() finds a symlink where it expected a directory it needs to return an error code that can be converted to NOTDIR for v2 or v3 but will be SYMLINK for v4. It must be different from the error code returns when it finds a symlink but expects a regular file - that must be converted to EINVAL or SYMLINK. So we introduce an internal error code nfserr_symlink_not_dir which each version converts as appropriate. nfsd_check_obj_isreg() is similar to nfsd_mode_check() except that it is only used by NFSv4 and only for OPEN. NFSERR_INVAL is never a suitable error if the object is the wrong time. For v4.0 we use nfserr_symlink for non-dirs even if not a symlink. For v4.1 we have nfserr_wrong_type. We handle this difference in-place in nfsd_check_obj_isreg() as there is nothing to be gained by delaying the choice to nfsd4_map_status(). As a result of these changes, nfsd_mode_check() doesn't need an rqstp arg any more. Note that NFSv4 operations are actually performed in the xdr code(!!!) so to the only place that we can map the status code successfully is in nfsd4_encode_operation(). Signed-off-by: NeilBrown <neilb@suse.de> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-08-14 21:21:01 +08:00
if (S_ISLNK(mode))
return nfserr_symlink;
/* RFC 7530 - 16.16.6 */
if (minor_version == 0)
return nfserr_symlink;
else
return nfserr_wrong_type;
nfsd4: return nfserr_symlink on v4 OPEN of non-regular file Without this, an attempt to open a device special file without first stat'ing it will fail. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-16 04:55:02 +08:00
}
nfsd4: more sessions/open-owner-replay cleanup More logic that's unnecessary in the 4.1 case. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-03-23 06:03:49 +08:00
static void nfsd4_set_open_owner_reply_cache(struct nfsd4_compound_state *cstate, struct nfsd4_open *open, struct svc_fh *resfh)
{
if (nfsd4_has_session(cstate))
return;
fh_copy_shallow(&open->op_openowner->oo_owner.so_replay.rp_openfh,
&resfh->fh_handle);
}
NFSD: Refactor NFSv4 OPEN(CREATE) Copy do_nfsd_create() to nfs4proc.c and remove NFSv3-specific logic. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-03-29 02:47:34 +08:00
static inline bool nfsd4_create_is_exclusive(int createmode)
{
return createmode == NFS4_CREATE_EXCLUSIVE ||
createmode == NFS4_CREATE_EXCLUSIVE4_1;
}
NFSD: Instantiate a struct file when creating a regular NFSv4 file There have been reports of races that cause NFSv4 OPEN(CREATE) to return an error even though the requested file was created. NFSv4 does not provide a status code for this case. To mitigate some of these problems, reorganize the NFSv4 OPEN(CREATE) logic to allocate resources before the file is actually created, and open the new file while the parent directory is still locked. Two new APIs are added: + Add an API that works like nfsd_file_acquire() but does not open the underlying file. The OPEN(CREATE) path can use this API when it already has an open file. + Add an API that is kin to dentry_open(). NFSD needs to create a file and grab an open "struct file *" atomically. The alloc_empty_file() has to be done before the inode create. If it fails (for example, because the NFS server has exceeded its max_files limit), we avoid creating the file and can still return an error to the NFS client. BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=382 Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Tested-by: JianHong Yin <jiyin@redhat.com>
2022-03-30 22:30:54 +08:00
static __be32
nfsd4_vfs_create(struct svc_fh *fhp, struct dentry *child,
struct nfsd4_open *open)
{
struct file *filp;
struct path path;
int oflags;
oflags = O_CREAT | O_LARGEFILE;
switch (open->op_share_access & NFS4_SHARE_ACCESS_BOTH) {
case NFS4_SHARE_ACCESS_WRITE:
oflags |= O_WRONLY;
break;
case NFS4_SHARE_ACCESS_BOTH:
oflags |= O_RDWR;
break;
default:
oflags |= O_RDONLY;
}
path.mnt = fhp->fh_export->ex_path.mnt;
path.dentry = child;
filp = dentry_create(&path, oflags, open->op_iattr.ia_mode,
current_cred());
if (IS_ERR(filp))
return nfserrno(PTR_ERR(filp));
open->op_filp = filp;
return nfs_ok;
}
NFSD: Refactor NFSv4 OPEN(CREATE) Copy do_nfsd_create() to nfs4proc.c and remove NFSv3-specific logic. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-03-29 02:47:34 +08:00
/*
* Implement NFSv4's unchecked, guarded, and exclusive create
* semantics for regular files. Open state for this new file is
* subsequently fabricated in nfsd4_process_open2().
*
* Upon return, caller must release @fhp and @resfhp.
*/
static __be32
nfsd4_create_file(struct svc_rqst *rqstp, struct svc_fh *fhp,
struct svc_fh *resfhp, struct nfsd4_open *open)
{
struct iattr *iap = &open->op_iattr;
NFSD: introduce struct nfsd_attrs The attributes that nfsd might want to set on a file include 'struct iattr' as well as an ACL and security label. The latter two are passed around quite separately from the first, in part because they are only needed for NFSv4. This leads to some clumsiness in the code, such as the attributes NOT being set in nfsd_create_setattr(). We need to keep the directory locked until all attributes are set to ensure the file is never visibile without all its attributes. This need combined with the inconsistent handling of attributes leads to more clumsiness. As a first step towards tidying this up, introduce 'struct nfsd_attrs'. This is passed (by reference) to vfs.c functions that work with attributes, and is assembled by the various nfs*proc functions which call them. As yet only iattr is included, but future patches will expand this. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
struct nfsd_attrs attrs = {
.na_iattr = iap,
NFSD: add security label to struct nfsd_attrs nfsd_setattr() now sets a security label if provided, and nfsv4 provides it in the 'open' and 'create' paths and the 'setattr' path. If setting the label failed (including because the kernel doesn't support labels), an error field in 'struct nfsd_attrs' is set, and the caller can respond. The open/create callers clear FATTR4_WORD2_SECURITY_LABEL in the returned attr set in this case. The setattr caller returns the error. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
.na_seclabel = &open->op_label,
NFSD: introduce struct nfsd_attrs The attributes that nfsd might want to set on a file include 'struct iattr' as well as an ACL and security label. The latter two are passed around quite separately from the first, in part because they are only needed for NFSv4. This leads to some clumsiness in the code, such as the attributes NOT being set in nfsd_create_setattr(). We need to keep the directory locked until all attributes are set to ensure the file is never visibile without all its attributes. This need combined with the inconsistent handling of attributes leads to more clumsiness. As a first step towards tidying this up, introduce 'struct nfsd_attrs'. This is passed (by reference) to vfs.c functions that work with attributes, and is assembled by the various nfs*proc functions which call them. As yet only iattr is included, but future patches will expand this. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
};
NFSD: Refactor NFSv4 OPEN(CREATE) Copy do_nfsd_create() to nfs4proc.c and remove NFSv3-specific logic. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-03-29 02:47:34 +08:00
struct dentry *parent, *child;
__u32 v_mtime, v_atime;
struct inode *inode;
__be32 status;
int host_err;
if (isdotent(open->op_fname, open->op_fnamelen))
return nfserr_exist;
if (!(iap->ia_valid & ATTR_MODE))
iap->ia_mode = 0;
status = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_EXEC);
if (status != nfs_ok)
return status;
parent = fhp->fh_dentry;
inode = d_inode(parent);
host_err = fh_want_write(fhp);
if (host_err)
return nfserrno(host_err);
NFSD: add posix ACLs to struct nfsd_attrs pacl and dpacl pointers are added to struct nfsd_attrs, which requires that we have an nfsd_attrs_free() function to free them. Those nfsv4 functions that can set ACLs now set up these pointers based on the passed in NFSv4 ACL. nfsd_setattr() sets the acls as appropriate. Errors are handled as with security labels. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
if (is_create_with_attrs(open))
nfsd4_acl_to_attr(NF4REG, open->op_acl, &attrs);
NFSD: use explicit lock/unlock for directory ops When creating or unlinking a name in a directory use explicit inode_lock_nested() instead of fh_lock(), and explicit calls to fh_fill_pre_attrs() and fh_fill_post_attrs(). This is already done for renames, with lock_rename() as the explicit locking. Also move the 'fill' calls closer to the operation that might change the attributes. This way they are avoided on some error paths. For the v2-only code in nfsproc.c, the fill calls are not replaced as they aren't needed. Making the locking explicit will simplify proposed future changes to locking for directories. It also makes it easily visible exactly where pre/post attributes are used - not all callers of fh_lock() actually need the pre/post attributes. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
inode_lock_nested(inode, I_MUTEX_PARENT);
NFSD: Refactor NFSv4 OPEN(CREATE) Copy do_nfsd_create() to nfs4proc.c and remove NFSv3-specific logic. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-03-29 02:47:34 +08:00
child = lookup_one_len(open->op_fname, parent, open->op_fnamelen);
if (IS_ERR(child)) {
status = nfserrno(PTR_ERR(child));
goto out;
}
if (d_really_is_negative(child)) {
status = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_CREATE);
if (status != nfs_ok)
goto out;
}
status = fh_compose(resfhp, fhp->fh_export, child, fhp);
if (status != nfs_ok)
goto out;
v_mtime = 0;
v_atime = 0;
if (nfsd4_create_is_exclusive(open->op_createmode)) {
u32 *verifier = (u32 *)open->op_verf.data;
/*
* Solaris 7 gets confused (bugid 4218508) if these have
* the high bit set, as do xfs filesystems without the
* "bigtime" feature. So just clear the high bits. If this
* is ever changed to use different attrs for storing the
* verifier, then do_open_lookup() will also need to be
* fixed accordingly.
*/
v_mtime = verifier[0] & 0x7fffffff;
v_atime = verifier[1] & 0x7fffffff;
}
if (d_really_is_positive(child)) {
NFSD: reduce locking in nfsd_lookup() nfsd_lookup() takes an exclusive lock on the parent inode, but no callers want the lock and it may not be needed at all if the result is in the dcache. Change nfsd_lookup_dentry() to not take the lock, and call lookup_one_len_locked() which takes lock only if needed. nfsd4_open() currently expects the lock to still be held, but that isn't necessary as nfsd_validate_delegated_dentry() provides required guarantees without the lock. NOTE: NFSv4 requires directory changeinfo for OPEN even when a create wasn't requested and no change happened. Now that nfsd_lookup() doesn't use fh_lock(), we need to explicitly fill the attributes when no create happens. A new fh_fill_both_attrs() is provided for that task. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
/* NFSv4 protocol requires change attributes even though
* no change happened.
*/
nfsd: handle failure to collect pre/post-op attrs more sanely Collecting pre_op_attrs can fail, in which case it's probably best to fail the whole operation. Change fh_fill_pre_attrs and fh_fill_both_attrs to return __be32, and have the callers check the return code and abort the operation if it's not nfs_ok. Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-07-21 22:29:10 +08:00
status = fh_fill_both_attrs(fhp);
if (status != nfs_ok)
goto out;
NFSD: reduce locking in nfsd_lookup() nfsd_lookup() takes an exclusive lock on the parent inode, but no callers want the lock and it may not be needed at all if the result is in the dcache. Change nfsd_lookup_dentry() to not take the lock, and call lookup_one_len_locked() which takes lock only if needed. nfsd4_open() currently expects the lock to still be held, but that isn't necessary as nfsd_validate_delegated_dentry() provides required guarantees without the lock. NOTE: NFSv4 requires directory changeinfo for OPEN even when a create wasn't requested and no change happened. Now that nfsd_lookup() doesn't use fh_lock(), we need to explicitly fill the attributes when no create happens. A new fh_fill_both_attrs() is provided for that task. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
NFSD: Refactor NFSv4 OPEN(CREATE) Copy do_nfsd_create() to nfs4proc.c and remove NFSv3-specific logic. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-03-29 02:47:34 +08:00
switch (open->op_createmode) {
case NFS4_CREATE_UNCHECKED:
if (!d_is_reg(child))
break;
/*
* In NFSv4, we don't want to truncate the file
* now. This would be wrong if the OPEN fails for
* some other reason. Furthermore, if the size is
* nonzero, we should ignore it according to spec!
*/
open->op_truncate = (iap->ia_valid & ATTR_SIZE) &&
!iap->ia_size;
break;
case NFS4_CREATE_GUARDED:
status = nfserr_exist;
break;
case NFS4_CREATE_EXCLUSIVE:
nfsd: convert to new timestamp accessors Convert to using the new inode timestamp accessor functions. Signed-off-by: Jeff Layton <jlayton@kernel.org> Link: https://lore.kernel.org/r/20231004185347.80880-50-jlayton@kernel.org Signed-off-by: Christian Brauner <brauner@kernel.org>
2023-10-05 02:52:37 +08:00
if (inode_get_mtime_sec(d_inode(child)) == v_mtime &&
inode_get_atime_sec(d_inode(child)) == v_atime &&
NFSD: Refactor NFSv4 OPEN(CREATE) Copy do_nfsd_create() to nfs4proc.c and remove NFSv3-specific logic. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-03-29 02:47:34 +08:00
d_inode(child)->i_size == 0) {
open->op_created = true;
break; /* subtle */
}
status = nfserr_exist;
break;
case NFS4_CREATE_EXCLUSIVE4_1:
nfsd: convert to new timestamp accessors Convert to using the new inode timestamp accessor functions. Signed-off-by: Jeff Layton <jlayton@kernel.org> Link: https://lore.kernel.org/r/20231004185347.80880-50-jlayton@kernel.org Signed-off-by: Christian Brauner <brauner@kernel.org>
2023-10-05 02:52:37 +08:00
if (inode_get_mtime_sec(d_inode(child)) == v_mtime &&
inode_get_atime_sec(d_inode(child)) == v_atime &&
NFSD: Refactor NFSv4 OPEN(CREATE) Copy do_nfsd_create() to nfs4proc.c and remove NFSv3-specific logic. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-03-29 02:47:34 +08:00
d_inode(child)->i_size == 0) {
open->op_created = true;
goto set_attr; /* subtle */
}
status = nfserr_exist;
}
goto out;
}
if (!IS_POSIXACL(inode))
iap->ia_mode &= ~current_umask();
nfsd: handle failure to collect pre/post-op attrs more sanely Collecting pre_op_attrs can fail, in which case it's probably best to fail the whole operation. Change fh_fill_pre_attrs and fh_fill_both_attrs to return __be32, and have the callers check the return code and abort the operation if it's not nfs_ok. Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-07-21 22:29:10 +08:00
status = fh_fill_pre_attrs(fhp);
if (status != nfs_ok)
goto out;
NFSD: Instantiate a struct file when creating a regular NFSv4 file There have been reports of races that cause NFSv4 OPEN(CREATE) to return an error even though the requested file was created. NFSv4 does not provide a status code for this case. To mitigate some of these problems, reorganize the NFSv4 OPEN(CREATE) logic to allocate resources before the file is actually created, and open the new file while the parent directory is still locked. Two new APIs are added: + Add an API that works like nfsd_file_acquire() but does not open the underlying file. The OPEN(CREATE) path can use this API when it already has an open file. + Add an API that is kin to dentry_open(). NFSD needs to create a file and grab an open "struct file *" atomically. The alloc_empty_file() has to be done before the inode create. If it fails (for example, because the NFS server has exceeded its max_files limit), we avoid creating the file and can still return an error to the NFS client. BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=382 Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Tested-by: JianHong Yin <jiyin@redhat.com>
2022-03-30 22:30:54 +08:00
status = nfsd4_vfs_create(fhp, child, open);
if (status != nfs_ok)
NFSD: Refactor NFSv4 OPEN(CREATE) Copy do_nfsd_create() to nfs4proc.c and remove NFSv3-specific logic. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-03-29 02:47:34 +08:00
goto out;
open->op_created = true;
NFSD: use explicit lock/unlock for directory ops When creating or unlinking a name in a directory use explicit inode_lock_nested() instead of fh_lock(), and explicit calls to fh_fill_pre_attrs() and fh_fill_post_attrs(). This is already done for renames, with lock_rename() as the explicit locking. Also move the 'fill' calls closer to the operation that might change the attributes. This way they are avoided on some error paths. For the v2-only code in nfsproc.c, the fill calls are not replaced as they aren't needed. Making the locking explicit will simplify proposed future changes to locking for directories. It also makes it easily visible exactly where pre/post attributes are used - not all callers of fh_lock() actually need the pre/post attributes. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
fh_fill_post_attrs(fhp);
NFSD: Refactor NFSv4 OPEN(CREATE) Copy do_nfsd_create() to nfs4proc.c and remove NFSv3-specific logic. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-03-29 02:47:34 +08:00
/* A newly created file already has a file size of zero. */
if ((iap->ia_valid & ATTR_SIZE) && (iap->ia_size == 0))
iap->ia_valid &= ~ATTR_SIZE;
if (nfsd4_create_is_exclusive(open->op_createmode)) {
iap->ia_valid = ATTR_MTIME | ATTR_ATIME |
ATTR_MTIME_SET|ATTR_ATIME_SET;
iap->ia_mtime.tv_sec = v_mtime;
iap->ia_atime.tv_sec = v_atime;
iap->ia_mtime.tv_nsec = 0;
iap->ia_atime.tv_nsec = 0;
}
set_attr:
NFSD: introduce struct nfsd_attrs The attributes that nfsd might want to set on a file include 'struct iattr' as well as an ACL and security label. The latter two are passed around quite separately from the first, in part because they are only needed for NFSv4. This leads to some clumsiness in the code, such as the attributes NOT being set in nfsd_create_setattr(). We need to keep the directory locked until all attributes are set to ensure the file is never visibile without all its attributes. This need combined with the inconsistent handling of attributes leads to more clumsiness. As a first step towards tidying this up, introduce 'struct nfsd_attrs'. This is passed (by reference) to vfs.c functions that work with attributes, and is assembled by the various nfs*proc functions which call them. As yet only iattr is included, but future patches will expand this. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
status = nfsd_create_setattr(rqstp, fhp, resfhp, &attrs);
NFSD: Refactor NFSv4 OPEN(CREATE) Copy do_nfsd_create() to nfs4proc.c and remove NFSv3-specific logic. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-03-29 02:47:34 +08:00
NFSD: add security label to struct nfsd_attrs nfsd_setattr() now sets a security label if provided, and nfsv4 provides it in the 'open' and 'create' paths and the 'setattr' path. If setting the label failed (including because the kernel doesn't support labels), an error field in 'struct nfsd_attrs' is set, and the caller can respond. The open/create callers clear FATTR4_WORD2_SECURITY_LABEL in the returned attr set in this case. The setattr caller returns the error. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
if (attrs.na_labelerr)
open->op_bmval[2] &= ~FATTR4_WORD2_SECURITY_LABEL;
NFSD: add posix ACLs to struct nfsd_attrs pacl and dpacl pointers are added to struct nfsd_attrs, which requires that we have an nfsd_attrs_free() function to free them. Those nfsv4 functions that can set ACLs now set up these pointers based on the passed in NFSv4 ACL. nfsd_setattr() sets the acls as appropriate. Errors are handled as with security labels. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
if (attrs.na_aclerr)
open->op_bmval[0] &= ~FATTR4_WORD0_ACL;
NFSD: Refactor NFSv4 OPEN(CREATE) Copy do_nfsd_create() to nfs4proc.c and remove NFSv3-specific logic. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-03-29 02:47:34 +08:00
out:
NFSD: use explicit lock/unlock for directory ops When creating or unlinking a name in a directory use explicit inode_lock_nested() instead of fh_lock(), and explicit calls to fh_fill_pre_attrs() and fh_fill_post_attrs(). This is already done for renames, with lock_rename() as the explicit locking. Also move the 'fill' calls closer to the operation that might change the attributes. This way they are avoided on some error paths. For the v2-only code in nfsproc.c, the fill calls are not replaced as they aren't needed. Making the locking explicit will simplify proposed future changes to locking for directories. It also makes it easily visible exactly where pre/post attributes are used - not all callers of fh_lock() actually need the pre/post attributes. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
inode_unlock(inode);
NFSD: add posix ACLs to struct nfsd_attrs pacl and dpacl pointers are added to struct nfsd_attrs, which requires that we have an nfsd_attrs_free() function to free them. Those nfsv4 functions that can set ACLs now set up these pointers based on the passed in NFSv4 ACL. nfsd_setattr() sets the acls as appropriate. Errors are handled as with security labels. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
nfsd_attrs_free(&attrs);
NFSD: Refactor NFSv4 OPEN(CREATE) Copy do_nfsd_create() to nfs4proc.c and remove NFSv3-specific logic. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-03-29 02:47:34 +08:00
if (child && !IS_ERR(child))
dput(child);
fh_drop_write(fhp);
return status;
}
nfsd: remove unsafe BUG_ON from set_change_info At one time, nfsd would scrape inode information directly out of struct inode in order to populate the change_info4. At that time, the BUG_ON in set_change_info made some sense, since having it unset meant a coding error. More recently, it calls vfs_getattr to get this information, which can fail. If that fails, fh_pre_saved can end up not being set. While this situation is unfortunate, we don't need to crash the box. Move set_change_info to nfs4proc.c since all of the callers are there. Revise the condition for setting "atomic" to also check for fh_pre_saved. Drop the BUG_ON and just have it zero out both change_attr4s when this occurs. Reported-by: Boyang Xue <bxue@redhat.com> Closes: https://bugzilla.redhat.com/show_bug.cgi?id=2223560 Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-07-21 22:29:11 +08:00
/**
* set_change_info - set up the change_info4 for a reply
* @cinfo: pointer to nfsd4_change_info to be populated
* @fhp: pointer to svc_fh to use as source
*
* Many operations in NFSv4 require change_info4 in the reply. This function
* populates that from the info that we (should!) have already collected. In
* the event that we didn't get any pre-attrs, just zero out both.
*/
static void
set_change_info(struct nfsd4_change_info *cinfo, struct svc_fh *fhp)
{
cinfo->atomic = (u32)(fhp->fh_pre_saved && fhp->fh_post_saved && !fhp->fh_no_atomic_attr);
cinfo->before_change = fhp->fh_pre_change;
cinfo->after_change = fhp->fh_post_change;
/*
* If fetching the pre-change attributes failed, then we should
* have already failed the whole operation. We could have still
* failed to fetch post-change attributes however.
*
* If we didn't get post-op attrs, just zero-out the after
* field since we don't know what it should be. If the pre_saved
* field isn't set for some reason, throw warning and just copy
* whatever is in the after field.
*/
if (WARN_ON_ONCE(!fhp->fh_pre_saved))
cinfo->before_change = 0;
if (!fhp->fh_post_saved)
nfsd: set missing after_change as before_change + 1 In the event that we can't fetch post_op_attr attributes, we still need to set a value for the after_change. The operation has already happened, so we're not able to return an error at that point, but we do want to ensure that the client knows that its cache should be invalidated. If we weren't able to fetch post-op attrs, then just set the after_change to before_change + 1. The atomic flag should already be clear in this case. Suggested-by: Neil Brown <neilb@suse.de> Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-07-24 22:53:39 +08:00
cinfo->after_change = cinfo->before_change + 1;
nfsd: remove unsafe BUG_ON from set_change_info At one time, nfsd would scrape inode information directly out of struct inode in order to populate the change_info4. At that time, the BUG_ON in set_change_info made some sense, since having it unset meant a coding error. More recently, it calls vfs_getattr to get this information, which can fail. If that fails, fh_pre_saved can end up not being set. While this situation is unfortunate, we don't need to crash the box. Move set_change_info to nfs4proc.c since all of the callers are there. Revise the condition for setting "atomic" to also check for fh_pre_saved. Drop the BUG_ON and just have it zero out both change_attr4s when this occurs. Reported-by: Boyang Xue <bxue@redhat.com> Closes: https://bugzilla.redhat.com/show_bug.cgi?id=2223560 Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-07-21 22:29:11 +08:00
}
[PATCH] nfsd: nfs4 code returns error values in net-endian Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:28:59 +08:00
static __be32
nfsd4: delay setting current_fh in open This is basically a no-op, to simplify a following patch. Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-01-28 06:26:06 +08:00
do_open_lookup(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_open *open, struct svc_fh **resfh)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
nfsd4: more sessions/open-owner-replay cleanup More logic that's unnecessary in the 4.1 case. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-03-23 06:03:49 +08:00
struct svc_fh *current_fh = &cstate->current_fh;
nfsd: avoid permission checks on EXCLUSIVE_CREATE replay With NFSv4, if we create a file then open it we explicit avoid checking the permissions on the file during the open because the fact that we created it ensures we should be allow to open it (the create and the open should appear to be a single operation). However if the reply to an EXCLUSIVE create gets lots and the client resends the create, the current code will perform the permission check - because it doesn't realise that it did the open already.. This patch should fix this. Note that I haven't actually seen this cause a problem. I was just looking at the code trying to figure out a different EXCLUSIVE open related issue, and this looked wrong. (Fix confirmed with pynfs 4.0 test OPEN4--bfields) Cc: stable@kernel.org Signed-off-by: NeilBrown <neilb@suse.de> [bfields: use OWNER_OVERRIDE and update for 4.1] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-12-08 04:40:55 +08:00
int accmode;
[PATCH] nfsd: nfs4 code returns error values in net-endian Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:28:59 +08:00
__be32 status;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
nfsd4: delay setting current_fh in open This is basically a no-op, to simplify a following patch. Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-01-28 06:26:06 +08:00
*resfh = kmalloc(sizeof(struct svc_fh), GFP_KERNEL);
if (!*resfh)
nfsd4: reduce do_open_lookup() stack usage I get 320 bytes for struct svc_fh on x86_64, really a little large to be putting on the stack; kmalloc() instead. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-01-28 05:49:55 +08:00
return nfserr_jukebox;
nfsd4: delay setting current_fh in open This is basically a no-op, to simplify a following patch. Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-01-28 06:26:06 +08:00
fh_init(*resfh, NFS4_FHSIZE);
nfsd: use true,false for bool variable in nfs4proc.c Fixes coccicheck warning: fs/nfsd/nfs4proc.c:235:1-18: WARNING: Assignment of 0/1 to bool variable fs/nfsd/nfs4proc.c:368:1-17: WARNING: Assignment of 0/1 to bool variable Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: zhengbin <zhengbin13@huawei.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-12-25 11:19:35 +08:00
open->op_truncate = false;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (open->op_create) {
nfsd41: CREATE_EXCLUSIVE4_1 Implement the CREATE_EXCLUSIVE4_1 open mode conforming to http://tools.ietf.org/html/draft-ietf-nfsv4-minorversion1-26 This mode allows the client to atomically create a file if it doesn't exist while setting some of its attributes. It must be implemented if the server supports persistent reply cache and/or pnfs. Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:29:17 +08:00
/* FIXME: check session persistence and pnfs flags.
* The nfsv4.1 spec requires the following semantics:
*
* Persistent | pNFS | Server REQUIRED | Client Allowed
* Reply Cache | server | |
* -------------+--------+-----------------+--------------------
* no | no | EXCLUSIVE4_1 | EXCLUSIVE4_1
* | | | (SHOULD)
* | | and EXCLUSIVE4 | or EXCLUSIVE4
* | | | (SHOULD NOT)
* no | yes | EXCLUSIVE4_1 | EXCLUSIVE4_1
* yes | no | GUARDED4 | GUARDED4
* yes | yes | GUARDED4 | GUARDED4
*/
nfsd: fix incorrect umasks We're neglecting to clear the umask after it's set, which can cause a later unrelated rpc to (incorrectly) use the same umask if it happens to be processed by the same thread. There's a more subtle problem here too: An NFSv4 compound request is decoded all in one pass before any operations are executed. Currently we're setting current->fs->umask at the time we decode the compound. In theory a single compound could contain multiple creates each setting a umask. In that case we'd end up using whichever umask was passed in the *last* operation as the umask for all the creates, whether that was correct or not. So, we should just be saving the umask at decode time and waiting to set it until we actually process the corresponding operation. In practice it's unlikely any client would do multiple creates in a single compound. And even if it did they'd likely be from the same process (hence carry the same umask). So this is a little academic, but we should get it right anyway. Fixes: 47057abde515 (nfsd: add support for the umask attribute) Cc: stable@vger.kernel.org Reported-by: Lucash Stach <l.stach@pengutronix.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-03-22 05:19:02 +08:00
current->fs->umask = open->op_umask;
NFSD: Refactor NFSv4 OPEN(CREATE) Copy do_nfsd_create() to nfs4proc.c and remove NFSv3-specific logic. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-03-29 02:47:34 +08:00
status = nfsd4_create_file(rqstp, current_fh, *resfh, open);
nfsd: fix incorrect umasks We're neglecting to clear the umask after it's set, which can cause a later unrelated rpc to (incorrectly) use the same umask if it happens to be processed by the same thread. There's a more subtle problem here too: An NFSv4 compound request is decoded all in one pass before any operations are executed. Currently we're setting current->fs->umask at the time we decode the compound. In theory a single compound could contain multiple creates each setting a umask. In that case we'd end up using whichever umask was passed in the *last* operation as the umask for all the creates, whether that was correct or not. So, we should just be saving the umask at decode time and waiting to set it until we actually process the corresponding operation. In practice it's unlikely any client would do multiple creates in a single compound. And even if it did they'd likely be from the same process (hence carry the same umask). So this is a little academic, but we should get it right anyway. Fixes: 47057abde515 (nfsd: add support for the umask attribute) Cc: stable@vger.kernel.org Reported-by: Lucash Stach <l.stach@pengutronix.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-03-22 05:19:02 +08:00
current->fs->umask = 0;
knfsd: set the response bitmask for NFS4_CREATE_EXCLUSIVE RFC 3530 says: If the server uses an attribute to store the exclusive create verifier, it will signify which attribute by setting the appropriate bit in the attribute mask that is returned in the results. Linux uses the atime and mtime to store the verifier, but sends a zeroed out bitmask back to the client. This patch makes sure that we set the correct bits in the bitmask in this situation. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Cc: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-31 15:37:51 +08:00
nfsd: clarify exclusive create bitmask result. The use of |= is confusing--the bitmask is always initialized to zero in this case, so we're effectively just doing an assignment here. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-02-03 04:12:27 +08:00
/*
NFSD: Set the attributes used to store the verifier for EXCLUSIVE4_1 According to rfc5661 18.16.4, "If EXCLUSIVE4_1 was used, the client determines the attributes used for the verifier by comparing attrset with cva_attrs.attrmask;" So, EXCLUSIVE4_1 also needs those bitmask used to store the verifier. Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2015-07-30 21:55:02 +08:00
* Following rfc 3530 14.2.16, and rfc 5661 18.16.4
* use the returned bitmask to indicate which attributes
* we used to store the verifier:
knfsd: set the response bitmask for NFS4_CREATE_EXCLUSIVE RFC 3530 says: If the server uses an attribute to store the exclusive create verifier, it will signify which attribute by setting the appropriate bit in the attribute mask that is returned in the results. Linux uses the atime and mtime to store the verifier, but sends a zeroed out bitmask back to the client. This patch makes sure that we set the correct bits in the bitmask in this situation. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Cc: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-31 15:37:51 +08:00
*/
NFSD: Refactor NFSv4 OPEN(CREATE) Copy do_nfsd_create() to nfs4proc.c and remove NFSv3-specific logic. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-03-29 02:47:34 +08:00
if (nfsd4_create_is_exclusive(open->op_createmode) && status == 0)
NFSD: Set the attributes used to store the verifier for EXCLUSIVE4_1 According to rfc5661 18.16.4, "If EXCLUSIVE4_1 was used, the client determines the attributes used for the verifier by comparing attrset with cva_attrs.attrmask;" So, EXCLUSIVE4_1 also needs those bitmask used to store the verifier. Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2015-07-30 21:55:02 +08:00
open->op_bmval[1] |= (FATTR4_WORD1_TIME_ACCESS |
FATTR4_WORD1_TIME_MODIFY);
NFSD: reduce locking in nfsd_lookup() nfsd_lookup() takes an exclusive lock on the parent inode, but no callers want the lock and it may not be needed at all if the result is in the dcache. Change nfsd_lookup_dentry() to not take the lock, and call lookup_one_len_locked() which takes lock only if needed. nfsd4_open() currently expects the lock to still be held, but that isn't necessary as nfsd_validate_delegated_dentry() provides required guarantees without the lock. NOTE: NFSv4 requires directory changeinfo for OPEN even when a create wasn't requested and no change happened. Now that nfsd_lookup() doesn't use fh_lock(), we need to explicitly fill the attributes when no create happens. A new fh_fill_both_attrs() is provided for that task. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
} else {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
status = nfsd_lookup(rqstp, current_fh,
NFSD: Add helper to decode OPEN's open_claim4 argument Refactor for clarity. Note that op_fname is the only instance of an NFSv4 filename stored in a struct xdr_netobj. Convert it to a u32/char * pair so that the new nfsd4_decode_filename() helper can be used. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-11-17 06:45:04 +08:00
open->op_fname, open->op_fnamelen, *resfh);
nfsd: handle failure to collect pre/post-op attrs more sanely Collecting pre_op_attrs can fail, in which case it's probably best to fail the whole operation. Change fh_fill_pre_attrs and fh_fill_both_attrs to return __be32, and have the callers check the return code and abort the operation if it's not nfs_ok. Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-07-21 22:29:10 +08:00
if (status == nfs_ok)
NFSD: reduce locking in nfsd_lookup() nfsd_lookup() takes an exclusive lock on the parent inode, but no callers want the lock and it may not be needed at all if the result is in the dcache. Change nfsd_lookup_dentry() to not take the lock, and call lookup_one_len_locked() which takes lock only if needed. nfsd4_open() currently expects the lock to still be held, but that isn't necessary as nfsd_validate_delegated_dentry() provides required guarantees without the lock. NOTE: NFSv4 requires directory changeinfo for OPEN even when a create wasn't requested and no change happened. Now that nfsd_lookup() doesn't use fh_lock(), we need to explicitly fill the attributes when no create happens. A new fh_fill_both_attrs() is provided for that task. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
/* NFSv4 protocol requires change attributes even though
* no change happened.
*/
nfsd: handle failure to collect pre/post-op attrs more sanely Collecting pre_op_attrs can fail, in which case it's probably best to fail the whole operation. Change fh_fill_pre_attrs and fh_fill_both_attrs to return __be32, and have the callers check the return code and abort the operation if it's not nfs_ok. Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-07-21 22:29:10 +08:00
status = fh_fill_both_attrs(current_fh);
NFSD: reduce locking in nfsd_lookup() nfsd_lookup() takes an exclusive lock on the parent inode, but no callers want the lock and it may not be needed at all if the result is in the dcache. Change nfsd_lookup_dentry() to not take the lock, and call lookup_one_len_locked() which takes lock only if needed. nfsd4_open() currently expects the lock to still be held, but that isn't necessary as nfsd_validate_delegated_dentry() provides required guarantees without the lock. NOTE: NFSv4 requires directory changeinfo for OPEN even when a create wasn't requested and no change happened. Now that nfsd_lookup() doesn't use fh_lock(), we need to explicitly fill the attributes when no create happens. A new fh_fill_both_attrs() is provided for that task. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
}
nfsd: don't fail unchecked creates of non-special files Allow a v3 unchecked open of a non-regular file succeed as if it were a lookup; typically a client in such a case will want to fall back on a local open, so succeeding and giving it the filehandle is more useful than failing with nfserr_exist, which makes it appear that nothing at all exists by that name. Similarly for v4, on an open-create, return the same errors we would on an attempt to open a non-regular file, instead of returning nfserr_exist. This fixes a problem found doing a v4 open of a symlink with O_RDONLY|O_CREAT, which resulted in the current client returning EEXIST. Thanks also to Trond for analysis. Cc: stable@kernel.org Reported-by: Orion Poplawski <orion@cora.nwra.com> Tested-by: Orion Poplawski <orion@cora.nwra.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-04-10 06:06:49 +08:00
if (status)
goto out;
nfsd: move error choice for incorrect object types to version-specific code. If an NFS operation expects a particular sort of object (file, dir, link, etc) but gets a file handle for a different sort of object, it must return an error. The actual error varies among NFS versions in non-trivial ways. For v2 and v3 there are ISDIR and NOTDIR errors and, for NFSv4 only, INVAL is suitable. For v4.0 there is also NFS4ERR_SYMLINK which should be used if a SYMLINK was found when not expected. This take precedence over NOTDIR. For v4.1+ there is also NFS4ERR_WRONG_TYPE which should be used in preference to EINVAL when none of the specific error codes apply. When nfsd_mode_check() finds a symlink where it expected a directory it needs to return an error code that can be converted to NOTDIR for v2 or v3 but will be SYMLINK for v4. It must be different from the error code returns when it finds a symlink but expects a regular file - that must be converted to EINVAL or SYMLINK. So we introduce an internal error code nfserr_symlink_not_dir which each version converts as appropriate. nfsd_check_obj_isreg() is similar to nfsd_mode_check() except that it is only used by NFSv4 and only for OPEN. NFSERR_INVAL is never a suitable error if the object is the wrong time. For v4.0 we use nfserr_symlink for non-dirs even if not a symlink. For v4.1 we have nfserr_wrong_type. We handle this difference in-place in nfsd_check_obj_isreg() as there is nothing to be gained by delaying the choice to nfsd4_map_status(). As a result of these changes, nfsd_mode_check() doesn't need an rqstp arg any more. Note that NFSv4 operations are actually performed in the xdr code(!!!) so to the only place that we can map the status code successfully is in nfsd4_encode_operation(). Signed-off-by: NeilBrown <neilb@suse.de> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-08-14 21:21:01 +08:00
status = nfsd_check_obj_isreg(*resfh, cstate->minorversion);
[PATCH] nfsd4: reindent do_open_lookup() Minor rearrangement, cleanup of do_open_lookup(). No change in behavior. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Acked-by: Neil Brown <neilb@suse.de> Cc: Jeff Garzik <jeff@garzik.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-11-09 09:44:39 +08:00
if (status)
goto out;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
nfsd4: delay setting current_fh in open This is basically a no-op, to simplify a following patch. Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-01-28 06:26:06 +08:00
nfsd4_set_open_owner_reply_cache(cstate, open, *resfh);
nfsd: avoid permission checks on EXCLUSIVE_CREATE replay With NFSv4, if we create a file then open it we explicit avoid checking the permissions on the file during the open because the fact that we created it ensures we should be allow to open it (the create and the open should appear to be a single operation). However if the reply to an EXCLUSIVE create gets lots and the client resends the create, the current code will perform the permission check - because it doesn't realise that it did the open already.. This patch should fix this. Note that I haven't actually seen this cause a problem. I was just looking at the code trying to figure out a different EXCLUSIVE open related issue, and this looked wrong. (Fix confirmed with pynfs 4.0 test OPEN4--bfields) Cc: stable@kernel.org Signed-off-by: NeilBrown <neilb@suse.de> [bfields: use OWNER_OVERRIDE and update for 4.1] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-12-08 04:40:55 +08:00
accmode = NFSD_MAY_NOP;
nfsd4: delegation-based open reclaims should bypass permissions We saw a v4.0 client's create fail as follows: - open create succeeds and gets a read delegation - client attempts to set mode on new file, gets DELAY while server recalls delegation. - client attempts a CLAIM_DELEGATE_CUR open using the delegation, gets error because of new file mode. This probably can't happen on a recent kernel since we're no longer giving out delegations on create opens. Nevertheless, it's a bug--reclaim opens should bypass permission checks. Reported-by: Steve Dickson <steved@redhat.com> Reported-by: Trond Myklebust <Trond.Myklebust@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-06-20 03:47:37 +08:00
if (open->op_created ||
open->op_claim_type == NFS4_OPEN_CLAIM_DELEGATE_CUR)
nfsd: avoid permission checks on EXCLUSIVE_CREATE replay With NFSv4, if we create a file then open it we explicit avoid checking the permissions on the file during the open because the fact that we created it ensures we should be allow to open it (the create and the open should appear to be a single operation). However if the reply to an EXCLUSIVE create gets lots and the client resends the create, the current code will perform the permission check - because it doesn't realise that it did the open already.. This patch should fix this. Note that I haven't actually seen this cause a problem. I was just looking at the code trying to figure out a different EXCLUSIVE open related issue, and this looked wrong. (Fix confirmed with pynfs 4.0 test OPEN4--bfields) Cc: stable@kernel.org Signed-off-by: NeilBrown <neilb@suse.de> [bfields: use OWNER_OVERRIDE and update for 4.1] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-12-08 04:40:55 +08:00
accmode |= NFSD_MAY_OWNER_OVERRIDE;
nfsd4: delay setting current_fh in open This is basically a no-op, to simplify a following patch. Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-01-28 06:26:06 +08:00
status = do_open_permission(rqstp, *resfh, open, accmode);
nfsd4: delay setting current filehandle till success Compound processing stops on error, so the current filehandle won't be used on error. Thus the order here doesn't really matter. It'll be more convenient to do it later, though. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-01-28 05:26:02 +08:00
set_change_info(&open->op_cinfo, current_fh);
[PATCH] nfsd4: reindent do_open_lookup() Minor rearrangement, cleanup of do_open_lookup(). No change in behavior. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Acked-by: Neil Brown <neilb@suse.de> Cc: Jeff Garzik <jeff@garzik.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-11-09 09:44:39 +08:00
out:
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return status;
}
[PATCH] nfsd: nfs4 code returns error values in net-endian Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:28:59 +08:00
static __be32
nfsd4: more sessions/open-owner-replay cleanup More logic that's unnecessary in the 4.1 case. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-03-23 06:03:49 +08:00
do_open_fhandle(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_open *open)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
nfsd4: more sessions/open-owner-replay cleanup More logic that's unnecessary in the 4.1 case. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-03-23 06:03:49 +08:00
struct svc_fh *current_fh = &cstate->current_fh;
nfsd4: don't allow owner override on 4.1 CLAIM_FH opens The Linux client is using CLAIM_FH to implement regular opens, not just recovery cases, so it depends on the server to check permissions correctly. Therefore the owner override, which may make sense in the delegation recovery case, isn't right in the CLAIM_FH case. Symptoms: on a client with 49f9a0fafd844c32f2abada047c0b9a5ba0d6255 "NFSv4.1: Enable open-by-filehandle", Bryan noticed this: touch test.txt chmod 000 test.txt echo test > test.txt succeeding. Cc: stable@kernel.org Reported-by: Bryan Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-05-04 04:09:09 +08:00
int accmode = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* We don't know the target directory, and therefore can not
* set the change info
*/
memset(&open->op_cinfo, 0, sizeof(struct nfsd4_change_info));
nfsd4: more sessions/open-owner-replay cleanup More logic that's unnecessary in the 4.1 case. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-03-23 06:03:49 +08:00
nfsd4_set_open_owner_reply_cache(cstate, open, current_fh);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
open->op_truncate = (open->op_iattr.ia_valid & ATTR_SIZE) &&
(open->op_iattr.ia_size == 0);
nfsd4: don't allow owner override on 4.1 CLAIM_FH opens The Linux client is using CLAIM_FH to implement regular opens, not just recovery cases, so it depends on the server to check permissions correctly. Therefore the owner override, which may make sense in the delegation recovery case, isn't right in the CLAIM_FH case. Symptoms: on a client with 49f9a0fafd844c32f2abada047c0b9a5ba0d6255 "NFSv4.1: Enable open-by-filehandle", Bryan noticed this: touch test.txt chmod 000 test.txt echo test > test.txt succeeding. Cc: stable@kernel.org Reported-by: Bryan Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-05-04 04:09:09 +08:00
/*
* In the delegation case, the client is telling us about an
* open that it *already* performed locally, some time ago. We
* should let it succeed now if possible.
*
* In the case of a CLAIM_FH open, on the other hand, the client
* may be counting on us to enforce permissions (the Linux 4.1
* client uses this for normal opens, for example).
*/
if (open->op_claim_type == NFS4_OPEN_CLAIM_DELEG_CUR_FH)
accmode = NFSD_MAY_OWNER_OVERRIDE;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
NFSD: remove redundant variable status Return value directly from fh_verify() do_open_permission() exp_pseudoroot() instead of getting value from redundant variable status. Reported-by: Zeal Robot <zealci@zte.com.cn> Signed-off-by: Jinpeng Cui <cui.jinpeng2@zte.com.cn> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-08-31 22:20:02 +08:00
return do_open_permission(rqstp, current_fh, open, accmode);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
nfsd41: clientid handling Extract the clientid from sessionid to set the op_clientid on open. Verify that the clid for other stateful ops is zero for minorversion != 0 Do all other checks for stateful ops without sessions. Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: Andy Adamson <andros@netapp.com> [fixed whitespace indent] Signed-off-by: Benny Halevy <bhalevy@panasas.com> [nfsd41 remove sl_session from nfsd4_open] Signed-off-by: Andy Adamson <andros@netapp.com> Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:28:50 +08:00
static void
copy_clientid(clientid_t *clid, struct nfsd4_session *session)
{
struct nfsd4_sessionid *sid =
(struct nfsd4_sessionid *)session->se_sessionid.data;
clid->cl_boot = sid->clientid.cl_boot;
clid->cl_id = sid->clientid.cl_id;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
[PATCH] knfsd: nfsd4: don't inline nfsd4 compound op functions The inlining contributes to bloating the stack of nfsd4_compound, and I want to change the compound op functions to function pointers anyway. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:29 +08:00
static __be32
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
nfsd4_open(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_open *open = &u->open;
[PATCH] nfsd: nfs4 code returns error values in net-endian Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:28:59 +08:00
__be32 status;
nfsd4: delay setting current_fh in open This is basically a no-op, to simplify a following patch. Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-01-28 06:26:06 +08:00
struct svc_fh *resfh = NULL;
nfsd: use service net instead of hard-coded init_net This patch replaces init_net by SVC_NET(), where possible and also passes proper context to nested functions where required. Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-11-14 23:22:07 +08:00
struct net *net = SVC_NET(rqstp);
struct nfsd_net *nn = net_generic(net, nfsd_net_id);
nfsd4: extend reclaim period for reclaiming clients If the client is only renewing state a little sooner than once a lease period, then it might not discover the server has restarted till close to the end of the grace period, and might run out of time to do the actual reclaim. Extend the grace period by a second each time we notice there are clients still trying to reclaim, up to a limit of another whole lease period. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-06-09 00:28:47 +08:00
bool reclaim = false;
nfsd41: stateid handling When sessions are used, stateful operation sequenceid and stateid handling are not used. When sessions are used, on the first open set the seqid to 1, mark state confirmed and skip seqid processing. When sessionas are used the stateid generation number is ignored when it is zero whereas without sessions bad_stateid or stale stateid is returned. Add flags to propagate session use to all stateful ops and down to check_stateid_generation. Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: Andy Adamson <andros@netapp.com> [nfsd4_has_session should return a boolean, not u32] Signed-off-by: Benny Halevy <bhalevy@panasas.com> [nfsd41: pass nfsd4_compoundres * to nfsd4_process_open1] [nfsd41: calculate HAS_SESSION in nfs4_preprocess_stateid_op] [nfsd41: calculate HAS_SESSION in nfs4_preprocess_seqid_op] Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:28:45 +08:00
nfsd4: split stateowners into open and lockowners The stateowner has some fields that only make sense for openowners, and some that only make sense for lockowners, and I find it a lot clearer if those are separated out. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-31 11:33:59 +08:00
dprintk("NFSD: nfsd4_open filename %.*s op_openowner %p\n",
NFSD: Add helper to decode OPEN's open_claim4 argument Refactor for clarity. Note that op_fname is the only instance of an NFSv4 filename stored in a struct xdr_netobj. Convert it to a u32/char * pair so that the new nfsd4_decode_filename() helper can be used. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-11-17 06:45:04 +08:00
(int)open->op_fnamelen, open->op_fname,
nfsd4: split stateowners into open and lockowners The stateowner has some fields that only make sense for openowners, and some that only make sense for lockowners, and I find it a lot clearer if those are separated out. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-31 11:33:59 +08:00
open->op_openowner);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
NFSD: Instantiate a struct file when creating a regular NFSv4 file There have been reports of races that cause NFSv4 OPEN(CREATE) to return an error even though the requested file was created. NFSv4 does not provide a status code for this case. To mitigate some of these problems, reorganize the NFSv4 OPEN(CREATE) logic to allocate resources before the file is actually created, and open the new file while the parent directory is still locked. Two new APIs are added: + Add an API that works like nfsd_file_acquire() but does not open the underlying file. The OPEN(CREATE) path can use this API when it already has an open file. + Add an API that is kin to dentry_open(). NFSD needs to create a file and grab an open "struct file *" atomically. The alloc_empty_file() has to be done before the inode create. If it fails (for example, because the NFS server has exceeded its max_files limit), we avoid creating the file and can still return an error to the NFS client. BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=382 Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Tested-by: JianHong Yin <jiyin@redhat.com>
2022-03-30 22:30:54 +08:00
open->op_filp = NULL;
NFSD: verify the opened dentry after setting a delegation Between opening a file and setting a delegation on it, someone could rename or unlink the dentry. If this happens, we do not want to grant a delegation on the open. On a CLAIM_NULL open, we're opening by filename, and we may (in the non-create case) or may not (in the create case) be holding i_rwsem when attempting to set a delegation. The latter case allows a race. After getting a lease, redo the lookup of the file being opened and validate that the resulting dentry matches the one in the open file description. To properly redo the lookup we need an rqst pointer to pass to nfsd_lookup_dentry(), so make sure that is available. Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
open->op_rqstp = rqstp;
NFSD: Instantiate a struct file when creating a regular NFSv4 file There have been reports of races that cause NFSv4 OPEN(CREATE) to return an error even though the requested file was created. NFSv4 does not provide a status code for this case. To mitigate some of these problems, reorganize the NFSv4 OPEN(CREATE) logic to allocate resources before the file is actually created, and open the new file while the parent directory is still locked. Two new APIs are added: + Add an API that works like nfsd_file_acquire() but does not open the underlying file. The OPEN(CREATE) path can use this API when it already has an open file. + Add an API that is kin to dentry_open(). NFSD needs to create a file and grab an open "struct file *" atomically. The alloc_empty_file() has to be done before the inode create. If it fails (for example, because the NFS server has exceeded its max_files limit), we avoid creating the file and can still return an error to the NFS client. BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=382 Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Tested-by: JianHong Yin <jiyin@redhat.com>
2022-03-30 22:30:54 +08:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* This check required by spec. */
if (open->op_create && open->op_claim_type != NFS4_OPEN_CLAIM_NULL)
return nfserr_inval;
nfsd: use true,false for bool variable in nfs4proc.c Fixes coccicheck warning: fs/nfsd/nfs4proc.c:235:1-18: WARNING: Assignment of 0/1 to bool variable fs/nfsd/nfs4proc.c:368:1-17: WARNING: Assignment of 0/1 to bool variable Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: zhengbin <zhengbin13@huawei.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-12-25 11:19:35 +08:00
open->op_created = false;
nfsd41: Deny new lock before RECLAIM_COMPLETE done Before nfs41 client's RECLAIM_COMPLETE done, nfs server should deny any new locks or opens. rfc5661: " Whenever a client establishes a new client ID and before it does the first non-reclaim operation that obtains a lock, it MUST send a RECLAIM_COMPLETE with rca_one_fs set to FALSE, even if there are no locks to reclaim. If non-reclaim locking operations are done before the RECLAIM_COMPLETE, an NFS4ERR_GRACE error will be returned. " Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-14 15:06:26 +08:00
/*
* RFC5661 18.51.3
* Before RECLAIM_COMPLETE done, server should deny new lock
*/
if (nfsd4_has_session(cstate) &&
nfsd: cstate->session->se_client -> cstate->clp I'm not sure why we're writing this out the hard way in so many places. Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2021-01-22 06:57:45 +08:00
!test_bit(NFSD4_CLIENT_RECLAIM_COMPLETE, &cstate->clp->cl_flags) &&
nfsd41: Deny new lock before RECLAIM_COMPLETE done Before nfs41 client's RECLAIM_COMPLETE done, nfs server should deny any new locks or opens. rfc5661: " Whenever a client establishes a new client ID and before it does the first non-reclaim operation that obtains a lock, it MUST send a RECLAIM_COMPLETE with rca_one_fs set to FALSE, even if there are no locks to reclaim. If non-reclaim locking operations are done before the RECLAIM_COMPLETE, an NFS4ERR_GRACE error will be returned. " Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-14 15:06:26 +08:00
open->op_claim_type != NFS4_OPEN_CLAIM_PREVIOUS)
return nfserr_grace;
nfsd41: clientid handling Extract the clientid from sessionid to set the op_clientid on open. Verify that the clid for other stateful ops is zero for minorversion != 0 Do all other checks for stateful ops without sessions. Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: Andy Adamson <andros@netapp.com> [fixed whitespace indent] Signed-off-by: Benny Halevy <bhalevy@panasas.com> [nfsd41 remove sl_session from nfsd4_open] Signed-off-by: Andy Adamson <andros@netapp.com> Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:28:50 +08:00
if (nfsd4_has_session(cstate))
copy_clientid(&open->op_clientid, cstate->session);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* check seqid for replay. set nfs4_owner */
nfsd: Remove unneeded values in nfsd4_open() Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2015-07-13 17:32:05 +08:00
status = nfsd4_process_open1(cstate, open, nn);
[PATCH] nfsd: nfs_replay_me We are using NFS_REPLAY_ME as a special error value that is never leaked to clients. That works fine; the only problem is mixing host- and network- endian values in the same objects. Network-endian equivalent would work just as fine; switch to it. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:29:03 +08:00
if (status == nfserr_replay_me) {
nfsd4: split stateowners into open and lockowners The stateowner has some fields that only make sense for openowners, and some that only make sense for lockowners, and I find it a lot clearer if those are separated out. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-31 11:33:59 +08:00
struct nfs4_replay *rp = &open->op_openowner->oo_owner.so_replay;
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
fh_put(&cstate->current_fh);
nfsd4: use helper for copying filehandles for replay Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-02-03 06:23:10 +08:00
fh_copy_shallow(&cstate->current_fh.fh_handle,
&rp->rp_openfh);
nfsd: rename MAY_ flags Rename nfsd_permission() specific MAY_* flags to NFSD_MAY_* to make it clear, that these are not used outside nfsd, and to avoid name and number space conflicts with the VFS. [comment from hch: rename MAY_READ, MAY_WRITE and MAY_EXEC as well] Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-06-16 19:20:29 +08:00
status = fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_NOP);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (status)
dprintk("nfsd4_open: replay failed"
" restoring previous filehandle\n");
else
[PATCH] nfsd: nfs_replay_me We are using NFS_REPLAY_ME as a special error value that is never leaked to clients. That works fine; the only problem is mixing host- and network- endian values in the same objects. Network-endian equivalent would work just as fine; switch to it. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:29:03 +08:00
status = nfserr_replay_me;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
if (status)
goto out;
nfsd4: handle seqid-mutating open errors from xdr decoding If a client sets an owner (or group_owner or acl) attribute on open for create, and the mapping of that owner to an id fails, then we return BAD_OWNER. But BAD_OWNER is a seqid-mutating error, so we can't shortcut the open processing that case: we have to at least look up the owner so we can find the seqid to bump. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-03-01 04:51:49 +08:00
if (open->op_xdr_error) {
status = open->op_xdr_error;
goto out;
}
[PATCH] nfsd4: don't create on open that fails due to ERR_GRACE In an earlier patch (commit b648330a1d741d5df8a5076b2a0a2519c69c8f41) I noted that a too-early grace-period check was preventing us from bumping the sequence id on open. Unfortunately in that patch I stupidly moved the grace-period check back too far, so now an open for create can succesfully create the file while still returning ERR_GRACE. The correct place for that check is after we've set the open_owner and handled any replays, but before we actually start mucking with the filesystem. Thanks to Avishay Traeger for reporting the bug. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-19 09:43:36 +08:00
NFSv4: do exact check about attribute specified Server should return NFS4ERR_ATTRNOTSUPP if an attribute specified is not supported in current environment. Operations CREATE, NVERIFY, OPEN, SETATTR and VERIFY should do this check. This bug is found when do newpynfs tests. The names of the tests that failed are following: CR12 NVF7a NVF7b NVF7c NVF7d NVF7f NVF7r NVF7s OPEN15 VF7a VF7b VF7c VF7d VF7f VF7r VF7s Add function do_check_fattr() to do exact check: 1, Check attribute specified is supported by the NFSv4 server or not. 2, Check FATTR4_WORD0_ACL & FATTR4_WORD0_FS_LOCATIONS are supported in current environment or not. 3, Check attribute specified is writable or not. step 1 and 3 are done in function nfsd4_decode_fattr() but removed to this function now. Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-05-16 16:22:31 +08:00
status = nfsd4_check_open_attributes(rqstp, cstate, open);
if (status)
goto out;
[PATCH] nfsd4: don't create on open that fails due to ERR_GRACE In an earlier patch (commit b648330a1d741d5df8a5076b2a0a2519c69c8f41) I noted that a too-early grace-period check was preventing us from bumping the sequence id on open. Unfortunately in that patch I stupidly moved the grace-period check back too far, so now an open for create can succesfully create the file while still returning ERR_GRACE. The correct place for that check is after we've set the open_owner and handled any replays, but before we actually start mucking with the filesystem. Thanks to Avishay Traeger for reporting the bug. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-19 09:43:36 +08:00
/* Openowner is now set, so sequence id will get bumped. Now we need
* these checks before we do any creates: */
[PATCH] knfsd: fix nfs4_open lock leak I just noticed that my patch "don't create on open that fails due to ERR_GRACE" (recently commited as fb553c0f17444e090db951b96df4d2d71b4f4b6b) had an obvious problem that causes a deadlock on reboot recovery. Sending in this now since it seems like a clear 2.6.16 candidate.--b. We're returning with a lock held in some error cases. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-08 04:58:32 +08:00
status = nfserr_grace;
lockd: NLM grace period shouldn't block NFSv4 opens NLM locks don't conflict with NFSv4 share reservations, so we're not going to learn anything new by watiting for them. They do conflict with NFSv4 locks and with delegations. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2015-08-07 00:47:02 +08:00
if (opens_in_grace(net) && open->op_claim_type != NFS4_OPEN_CLAIM_PREVIOUS)
[PATCH] knfsd: fix nfs4_open lock leak I just noticed that my patch "don't create on open that fails due to ERR_GRACE" (recently commited as fb553c0f17444e090db951b96df4d2d71b4f4b6b) had an obvious problem that causes a deadlock on reboot recovery. Sending in this now since it seems like a clear 2.6.16 candidate.--b. We're returning with a lock held in some error cases. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-08 04:58:32 +08:00
goto out;
status = nfserr_no_grace;
lockd: NLM grace period shouldn't block NFSv4 opens NLM locks don't conflict with NFSv4 share reservations, so we're not going to learn anything new by watiting for them. They do conflict with NFSv4 locks and with delegations. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2015-08-07 00:47:02 +08:00
if (!opens_in_grace(net) && open->op_claim_type == NFS4_OPEN_CLAIM_PREVIOUS)
[PATCH] knfsd: fix nfs4_open lock leak I just noticed that my patch "don't create on open that fails due to ERR_GRACE" (recently commited as fb553c0f17444e090db951b96df4d2d71b4f4b6b) had an obvious problem that causes a deadlock on reboot recovery. Sending in this now since it seems like a clear 2.6.16 candidate.--b. We're returning with a lock held in some error cases. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-08 04:58:32 +08:00
goto out;
[PATCH] nfsd4: don't create on open that fails due to ERR_GRACE In an earlier patch (commit b648330a1d741d5df8a5076b2a0a2519c69c8f41) I noted that a too-early grace-period check was preventing us from bumping the sequence id on open. Unfortunately in that patch I stupidly moved the grace-period check back too far, so now an open for create can succesfully create the file while still returning ERR_GRACE. The correct place for that check is after we've set the open_owner and handled any replays, but before we actually start mucking with the filesystem. Thanks to Avishay Traeger for reporting the bug. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-19 09:43:36 +08:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
switch (open->op_claim_type) {
NFSD: Fix whitespace Clean up: Pull case arms back one tab stop to conform every other switch statement in fs/nfsd/nfs4proc.c. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-03-22 04:41:32 +08:00
case NFS4_OPEN_CLAIM_DELEGATE_CUR:
case NFS4_OPEN_CLAIM_NULL:
status = do_open_lookup(rqstp, cstate, open, &resfh);
if (status)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
goto out;
NFSD: Fix whitespace Clean up: Pull case arms back one tab stop to conform every other switch statement in fs/nfsd/nfs4proc.c. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-03-22 04:41:32 +08:00
break;
case NFS4_OPEN_CLAIM_PREVIOUS:
status = nfs4_check_open_reclaim(cstate->clp);
if (status)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
goto out;
NFSD: Fix whitespace Clean up: Pull case arms back one tab stop to conform every other switch statement in fs/nfsd/nfs4proc.c. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-03-22 04:41:32 +08:00
open->op_openowner->oo_flags |= NFS4_OO_CONFIRMED;
reclaim = true;
fallthrough;
case NFS4_OPEN_CLAIM_FH:
case NFS4_OPEN_CLAIM_DELEG_CUR_FH:
status = do_open_fhandle(rqstp, cstate, open);
if (status)
goto out;
resfh = &cstate->current_fh;
break;
case NFS4_OPEN_CLAIM_DELEG_PREV_FH:
case NFS4_OPEN_CLAIM_DELEGATE_PREV:
status = nfserr_notsupp;
goto out;
default:
status = nfserr_inval;
goto out;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
NFSD: Move documenting comment for nfsd4_process_open2() Clean up nfsd4_open() by converting a large comment at the only call site for nfsd4_process_open2() to a kerneldoc comment in front of that function. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-03-24 01:55:37 +08:00
nfsd4: delay setting current_fh in open This is basically a no-op, to simplify a following patch. Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-01-28 06:26:06 +08:00
status = nfsd4_process_open2(rqstp, resfh, open);
NFSD: Demote a WARN to a pr_warn() The call trace doesn't add much value, but it sure is noisy. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-09 02:23:45 +08:00
if (status && open->op_created)
pr_warn("nfsd4_process_open2 failed to open newly-created file: status=%u\n",
be32_to_cpu(status));
nfsd4: extend reclaim period for reclaiming clients If the client is only renewing state a little sooner than once a lease period, then it might not discover the server has restarted till close to the end of the grace period, and might run out of time to do the actual reclaim. Extend the grace period by a second each time we notice there are clients still trying to reclaim, up to a limit of another whole lease period. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-06-09 00:28:47 +08:00
if (reclaim && !status)
nn->somebody_reclaimed = true;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
out:
NFSD: Instantiate a struct file when creating a regular NFSv4 file There have been reports of races that cause NFSv4 OPEN(CREATE) to return an error even though the requested file was created. NFSv4 does not provide a status code for this case. To mitigate some of these problems, reorganize the NFSv4 OPEN(CREATE) logic to allocate resources before the file is actually created, and open the new file while the parent directory is still locked. Two new APIs are added: + Add an API that works like nfsd_file_acquire() but does not open the underlying file. The OPEN(CREATE) path can use this API when it already has an open file. + Add an API that is kin to dentry_open(). NFSD needs to create a file and grab an open "struct file *" atomically. The alloc_empty_file() has to be done before the inode create. If it fails (for example, because the NFS server has exceeded its max_files limit), we avoid creating the file and can still return an error to the NFS client. BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=382 Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Tested-by: JianHong Yin <jiyin@redhat.com>
2022-03-30 22:30:54 +08:00
if (open->op_filp) {
fput(open->op_filp);
open->op_filp = NULL;
}
nfsd4: delay setting current_fh in open This is basically a no-op, to simplify a following patch. Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-01-28 06:26:06 +08:00
if (resfh && resfh != &cstate->current_fh) {
fh_dup2(&cstate->current_fh, resfh);
fh_put(resfh);
kfree(resfh);
}
nfsd: remove unused status arg to nfsd4_cleanup_open_state Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2015-03-23 22:53:44 +08:00
nfsd4_cleanup_open_state(cstate, open);
nfsd4: cleanup handling of nfsv4.0 closed stateid's Closed stateid's are kept around a little while to handle close replays in the 4.0 case. So we stash them in the last-used stateid in the oo_last_closed_stateid field of the open owner. We can free that in encode_seqid_op_tail once the seqid on the open owner is next incremented. But we don't want to do that on the close itself; so we set NFS4_OO_PURGE_CLOSE flag set on the open owner, skip freeing it the first time through encode_seqid_op_tail, then when we see that flag set next time we free it. This is unnecessarily baroque. Instead, just move the logic that increments the seqid out of the xdr code and into the operation code itself. The justification given for the current placement is that we need to wait till the last minute to be sure we know whether the status is a sequence-id-mutating error or not, but examination of the code shows that can't actually happen. Reported-by: Yanchuan Nian <ycnian@gmail.com> Tested-by: Yanchuan Nian <ycnian@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-04-02 04:37:12 +08:00
nfsd4_bump_seqid(cstate, status);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return status;
}
nfsd4: handle seqid-mutating open errors from xdr decoding If a client sets an owner (or group_owner or acl) attribute on open for create, and the mapping of that owner to an id fails, then we return BAD_OWNER. But BAD_OWNER is a seqid-mutating error, so we can't shortcut the open processing that case: we have to at least look up the owner so we can find the seqid to bump. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-03-01 04:51:49 +08:00
/*
* OPEN is the only seqid-mutating operation whose decoding can fail
* with a seqid-mutating error (specifically, decoding of user names in
* the attributes). Therefore we have to do some processing to look up
* the stateowner so that we can bump the seqid.
*/
static __be32 nfsd4_open_omfg(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_op *op)
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_open *open = &op->u.open;
nfsd4: handle seqid-mutating open errors from xdr decoding If a client sets an owner (or group_owner or acl) attribute on open for create, and the mapping of that owner to an id fails, then we return BAD_OWNER. But BAD_OWNER is a seqid-mutating error, so we can't shortcut the open processing that case: we have to at least look up the owner so we can find the seqid to bump. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-03-01 04:51:49 +08:00
if (!seqid_mutating_err(ntohl(op->status)))
return op->status;
if (nfsd4_has_session(cstate))
return op->status;
open->op_xdr_error = op->status;
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
return nfsd4_open(rqstp, cstate, &op->u);
nfsd4: handle seqid-mutating open errors from xdr decoding If a client sets an owner (or group_owner or acl) attribute on open for create, and the mapping of that owner to an id fails, then we return BAD_OWNER. But BAD_OWNER is a seqid-mutating error, so we can't shortcut the open processing that case: we have to at least look up the owner so we can find the seqid to bump. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-03-01 04:51:49 +08:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/*
* filehandle-manipulating ops.
*/
[PATCH] knfsd: nfsd4: don't inline nfsd4 compound op functions The inlining contributes to bloating the stack of nfsd4_compound, and I want to change the compound op functions to function pointers anyway. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:29 +08:00
static __be32
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
nfsd4_getfh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
u->getfh = &cstate->current_fh;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return nfs_ok;
}
[PATCH] knfsd: nfsd4: don't inline nfsd4 compound op functions The inlining contributes to bloating the stack of nfsd4_compound, and I want to change the compound op functions to function pointers anyway. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:29 +08:00
static __be32
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
nfsd4_putfh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_putfh *putfh = &u->putfh;
NFSD: allow inter server COPY to have a STALE source server fh The inter server to server COPY source server filehandle is a foreign filehandle as the COPY is sent to the destination server. Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-07 22:56:48 +08:00
__be32 ret;
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
fh_put(&cstate->current_fh);
cstate->current_fh.fh_handle.fh_size = putfh->pf_fhlen;
NFSD: simplify struct nfsfh Most of the fields in 'struct knfsd_fh' are 2 levels deep (a union and a struct) and are accessed using macros like: #define fh_FOO fh_base.fh_new.fb_FOO This patch makes the union and struct anonymous, so that "fh_FOO" can be a name directly within 'struct knfsd_fh' and the #defines aren't needed. The file handle as a whole is sometimes accessed as "fh_base" or "fh_base.fh_pad", neither of which are particularly helpful names. As the struct holding the filehandle is now anonymous, we cannot use the name of that, so we union it with 'fh_raw' and use that where the raw filehandle is needed. fh_raw also ensure the structure is large enough for the largest possible filehandle. fh_raw is a 'char' array, removing any need to cast it for memcpy etc. SVCFH_fmt() is simplified using the "%ph" printk format. This changes the appearance of filehandles in dprintk() debugging, making them a little more precise. Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-09-02 09:16:32 +08:00
memcpy(&cstate->current_fh.fh_handle.fh_raw, putfh->pf_fhval,
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
putfh->pf_fhlen);
NFSD: allow inter server COPY to have a STALE source server fh The inter server to server COPY source server filehandle is a foreign filehandle as the COPY is sent to the destination server. Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-07 22:56:48 +08:00
ret = fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_BYPASS_GSS);
#ifdef CONFIG_NFSD_V4_2_INTER_SSC
if (ret == nfserr_stale && putfh->no_verify) {
SET_FH_FLAG(&cstate->current_fh, NFSD4_FH_FOREIGN);
ret = 0;
}
#endif
return ret;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
[PATCH] knfsd: nfsd4: don't inline nfsd4 compound op functions The inlining contributes to bloating the stack of nfsd4_compound, and I want to change the compound op functions to function pointers anyway. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:29 +08:00
static __be32
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
nfsd4_putrootfh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
fh_put(&cstate->current_fh);
NFSD: remove redundant variable status Return value directly from fh_verify() do_open_permission() exp_pseudoroot() instead of getting value from redundant variable status. Reported-by: Zeal Robot <zealci@zte.com.cn> Signed-off-by: Jinpeng Cui <cui.jinpeng2@zte.com.cn> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-08-31 22:20:02 +08:00
return exp_pseudoroot(rqstp, &cstate->current_fh);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
[PATCH] knfsd: nfsd4: don't inline nfsd4 compound op functions The inlining contributes to bloating the stack of nfsd4_compound, and I want to change the compound op functions to function pointers anyway. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:29 +08:00
static __be32
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
nfsd4_restorefh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
if (!cstate->save_fh.fh_dentry)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return nfserr_restorefh;
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
fh_dup2(&cstate->current_fh, &cstate->save_fh);
NFSD generalize nfsd4_compound_state flag names Allow for sid_flag field non-stateid use. Signed-off-by: Andy Adamson <andros@netapp.com>
2018-09-14 01:58:24 +08:00
if (HAS_CSTATE_FLAG(cstate, SAVED_STATE_ID_FLAG)) {
nfsd41: use current stateid by value Signed-off-by: Tigran Mkrtchyan <kofemann@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-02-14 05:55:32 +08:00
memcpy(&cstate->current_stateid, &cstate->save_stateid, sizeof(stateid_t));
NFSD generalize nfsd4_compound_state flag names Allow for sid_flag field non-stateid use. Signed-off-by: Andy Adamson <andros@netapp.com>
2018-09-14 01:58:24 +08:00
SET_CSTATE_FLAG(cstate, CURRENT_STATE_ID_FLAG);
nfsd41: use current stateid by value Signed-off-by: Tigran Mkrtchyan <kofemann@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-02-14 05:55:32 +08:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return nfs_ok;
}
[PATCH] knfsd: nfsd4: don't inline nfsd4 compound op functions The inlining contributes to bloating the stack of nfsd4_compound, and I want to change the compound op functions to function pointers anyway. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:29 +08:00
static __be32
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
nfsd4_savefh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
fh_dup2(&cstate->save_fh, &cstate->current_fh);
NFSD generalize nfsd4_compound_state flag names Allow for sid_flag field non-stateid use. Signed-off-by: Andy Adamson <andros@netapp.com>
2018-09-14 01:58:24 +08:00
if (HAS_CSTATE_FLAG(cstate, CURRENT_STATE_ID_FLAG)) {
nfsd41: use current stateid by value Signed-off-by: Tigran Mkrtchyan <kofemann@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-02-14 05:55:32 +08:00
memcpy(&cstate->save_stateid, &cstate->current_stateid, sizeof(stateid_t));
NFSD generalize nfsd4_compound_state flag names Allow for sid_flag field non-stateid use. Signed-off-by: Andy Adamson <andros@netapp.com>
2018-09-14 01:58:24 +08:00
SET_CSTATE_FLAG(cstate, SAVED_STATE_ID_FLAG);
nfsd41: use current stateid by value Signed-off-by: Tigran Mkrtchyan <kofemann@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-02-14 05:55:32 +08:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return nfs_ok;
}
/*
* misc nfsv4 ops
*/
[PATCH] knfsd: nfsd4: don't inline nfsd4 compound op functions The inlining contributes to bloating the stack of nfsd4_compound, and I want to change the compound op functions to function pointers anyway. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:29 +08:00
static __be32
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
nfsd4_access(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_access *access = &u->access;
nfsd: take xattr bits into account for permission checks Since the NFSv4.2 extended attributes extension defines 3 new access bits for xattr operations, take them in to account when validating what the client is asking for, and when checking permissions. Signed-off-by: Frank van der Linden <fllinden@amazon.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-06-24 06:39:24 +08:00
u32 access_full;
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
nfsd: take xattr bits into account for permission checks Since the NFSv4.2 extended attributes extension defines 3 new access bits for xattr operations, take them in to account when validating what the client is asking for, and when checking permissions. Signed-off-by: Frank van der Linden <fllinden@amazon.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-06-24 06:39:24 +08:00
access_full = NFS3_ACCESS_FULL;
if (cstate->minorversion >= 2)
access_full |= NFS4_ACCESS_XALIST | NFS4_ACCESS_XAREAD |
NFS4_ACCESS_XAWRITE;
if (access->ac_req_access & ~access_full)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return nfserr_inval;
access->ac_resp_access = access->ac_req_access;
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
return nfsd_access(rqstp, &cstate->current_fh, &access->ac_resp_access,
&access->ac_supported);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
[PATCH] knfsd: nfsd4: don't inline nfsd4 compound op functions The inlining contributes to bloating the stack of nfsd4_compound, and I want to change the compound op functions to function pointers anyway. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:29 +08:00
static __be32
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
nfsd4_commit(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_commit *commit = &u->commit;
NFSD: Pass the target nfsd_file to nfsd_commit() In a moment I'm going to introduce separate nfsd_file types, one of which is garbage-collected; the other, not. The garbage-collected variety is to be used by NFSv2 and v3, and the non-garbage-collected variety is to be used by NFSv4. nfsd_commit() is invoked by both NFSv3 and NFSv4 consumers. We want nfsd_commit() to find and use the correct variety of cached nfsd_file object for the NFS version that is in use. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Tested-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: NeilBrown <neilb@suse.de>
2022-10-28 22:46:38 +08:00
struct nfsd_file *nf;
__be32 status;
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
NFSD: Pass the target nfsd_file to nfsd_commit() In a moment I'm going to introduce separate nfsd_file types, one of which is garbage-collected; the other, not. The garbage-collected variety is to be used by NFSv2 and v3, and the non-garbage-collected variety is to be used by NFSv4. nfsd_commit() is invoked by both NFSv3 and NFSv4 consumers. We want nfsd_commit() to find and use the correct variety of cached nfsd_file object for the NFS version that is in use. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Tested-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: NeilBrown <neilb@suse.de>
2022-10-28 22:46:38 +08:00
status = nfsd_file_acquire(rqstp, &cstate->current_fh, NFSD_MAY_WRITE |
NFSD_MAY_NOT_BREAK_LEASE, &nf);
if (status != nfs_ok)
return status;
status = nfsd_commit(rqstp, &cstate->current_fh, nf, commit->co_offset,
nfsd: Ensure sampling of the commit verifier is atomic with the commit When we have a successful commit, ensure we sample the commit verifier before releasing the lock. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2020-01-07 02:40:36 +08:00
commit->co_count,
(__be32 *)commit->co_verf.data);
NFSD: Pass the target nfsd_file to nfsd_commit() In a moment I'm going to introduce separate nfsd_file types, one of which is garbage-collected; the other, not. The garbage-collected variety is to be used by NFSv2 and v3, and the non-garbage-collected variety is to be used by NFSv4. nfsd_commit() is invoked by both NFSv3 and NFSv4 consumers. We want nfsd_commit() to find and use the correct variety of cached nfsd_file object for the NFS version that is in use. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Tested-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: NeilBrown <neilb@suse.de>
2022-10-28 22:46:38 +08:00
nfsd_file_put(nf);
return status;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
[PATCH] nfsd: nfs4 code returns error values in net-endian Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:28:59 +08:00
static __be32
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
nfsd4_create(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_create *create = &u->create;
NFSD: introduce struct nfsd_attrs The attributes that nfsd might want to set on a file include 'struct iattr' as well as an ACL and security label. The latter two are passed around quite separately from the first, in part because they are only needed for NFSv4. This leads to some clumsiness in the code, such as the attributes NOT being set in nfsd_create_setattr(). We need to keep the directory locked until all attributes are set to ensure the file is never visibile without all its attributes. This need combined with the inconsistent handling of attributes leads to more clumsiness. As a first step towards tidying this up, introduce 'struct nfsd_attrs'. This is passed (by reference) to vfs.c functions that work with attributes, and is assembled by the various nfs*proc functions which call them. As yet only iattr is included, but future patches will expand this. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
struct nfsd_attrs attrs = {
.na_iattr = &create->cr_iattr,
NFSD: add security label to struct nfsd_attrs nfsd_setattr() now sets a security label if provided, and nfsv4 provides it in the 'open' and 'create' paths and the 'setattr' path. If setting the label failed (including because the kernel doesn't support labels), an error field in 'struct nfsd_attrs' is set, and the caller can respond. The open/create callers clear FATTR4_WORD2_SECURITY_LABEL in the returned attr set in this case. The setattr caller returns the error. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
.na_seclabel = &create->cr_label,
NFSD: introduce struct nfsd_attrs The attributes that nfsd might want to set on a file include 'struct iattr' as well as an ACL and security label. The latter two are passed around quite separately from the first, in part because they are only needed for NFSv4. This leads to some clumsiness in the code, such as the attributes NOT being set in nfsd_create_setattr(). We need to keep the directory locked until all attributes are set to ensure the file is never visibile without all its attributes. This need combined with the inconsistent handling of attributes leads to more clumsiness. As a first step towards tidying this up, introduce 'struct nfsd_attrs'. This is passed (by reference) to vfs.c functions that work with attributes, and is assembled by the various nfs*proc functions which call them. As yet only iattr is included, but future patches will expand this. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
};
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
struct svc_fh resfh;
[PATCH] nfsd: nfs4 code returns error values in net-endian Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:28:59 +08:00
__be32 status;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
dev_t rdev;
fh_init(&resfh, NFS4_FHSIZE);
nfsd: drop unnecessary MAY_EXEC check from create We need an fh_verify to make sure we at least have a dentry, but actual permission checks happen later. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-07-22 04:00:12 +08:00
status = fh_verify(rqstp, &cstate->current_fh, S_IFDIR, NFSD_MAY_NOP);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (status)
return status;
NFSv4: do exact check about attribute specified Server should return NFS4ERR_ATTRNOTSUPP if an attribute specified is not supported in current environment. Operations CREATE, NVERIFY, OPEN, SETATTR and VERIFY should do this check. This bug is found when do newpynfs tests. The names of the tests that failed are following: CR12 NVF7a NVF7b NVF7c NVF7d NVF7f NVF7r NVF7s OPEN15 VF7a VF7b VF7c VF7d VF7f VF7r VF7s Add function do_check_fattr() to do exact check: 1, Check attribute specified is supported by the NFSv4 server or not. 2, Check FATTR4_WORD0_ACL & FATTR4_WORD0_FS_LOCATIONS are supported in current environment or not. 3, Check attribute specified is writable or not. step 1 and 3 are done in function nfsd4_decode_fattr() but removed to this function now. Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-05-16 16:22:31 +08:00
status = check_attr_support(rqstp, cstate, create->cr_bmval,
nfsd_attrmask);
if (status)
return status;
NFSD: add posix ACLs to struct nfsd_attrs pacl and dpacl pointers are added to struct nfsd_attrs, which requires that we have an nfsd_attrs_free() function to free them. Those nfsv4 functions that can set ACLs now set up these pointers based on the passed in NFSv4 ACL. nfsd_setattr() sets the acls as appropriate. Errors are handled as with security labels. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
status = nfsd4_acl_to_attr(create->cr_type, create->cr_acl, &attrs);
nfsd: fix incorrect umasks We're neglecting to clear the umask after it's set, which can cause a later unrelated rpc to (incorrectly) use the same umask if it happens to be processed by the same thread. There's a more subtle problem here too: An NFSv4 compound request is decoded all in one pass before any operations are executed. Currently we're setting current->fs->umask at the time we decode the compound. In theory a single compound could contain multiple creates each setting a umask. In that case we'd end up using whichever umask was passed in the *last* operation as the umask for all the creates, whether that was correct or not. So, we should just be saving the umask at decode time and waiting to set it until we actually process the corresponding operation. In practice it's unlikely any client would do multiple creates in a single compound. And even if it did they'd likely be from the same process (hence carry the same umask). So this is a little academic, but we should get it right anyway. Fixes: 47057abde515 (nfsd: add support for the umask attribute) Cc: stable@vger.kernel.org Reported-by: Lucash Stach <l.stach@pengutronix.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-03-22 05:19:02 +08:00
current->fs->umask = create->cr_umask;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
switch (create->cr_type) {
case NF4LNK:
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
status = nfsd_symlink(rqstp, &cstate->current_fh,
create->cr_name, create->cr_namelen,
NFSD: set attributes when creating symlinks The NFS protocol includes attributes when creating symlinks. Linux does store attributes for symlinks and allows them to be set, though they are not used for permission checking. NFSD currently doesn't set standard (struct iattr) attributes when creating symlinks, but for NFSv4 it does set ACLs and security labels. This is inconsistent. To improve consistency, pass the provided attributes into nfsd_symlink() and call nfsd_create_setattr() to set them. NOTE: this results in a behaviour change for all NFS versions when the client sends non-default attributes with a SYMLINK request. With the Linux client, the only attributes are: attr.ia_mode = S_IFLNK | S_IRWXUGO; attr.ia_valid = ATTR_MODE; so the final outcome will be unchanged. Other clients might sent different attributes, and if they did they probably expect them to be honoured. We ignore any error from nfsd_create_setattr(). It isn't really clear what should be done if a file is successfully created, but the attributes cannot be set. NFS doesn't allow partial success to be reported. Reporting failure is probably more misleading than reporting success, so the status is ignored. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
create->cr_data, &attrs, &resfh);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
break;
case NF4BLK:
nfsd: fix incorrect umasks We're neglecting to clear the umask after it's set, which can cause a later unrelated rpc to (incorrectly) use the same umask if it happens to be processed by the same thread. There's a more subtle problem here too: An NFSv4 compound request is decoded all in one pass before any operations are executed. Currently we're setting current->fs->umask at the time we decode the compound. In theory a single compound could contain multiple creates each setting a umask. In that case we'd end up using whichever umask was passed in the *last* operation as the umask for all the creates, whether that was correct or not. So, we should just be saving the umask at decode time and waiting to set it until we actually process the corresponding operation. In practice it's unlikely any client would do multiple creates in a single compound. And even if it did they'd likely be from the same process (hence carry the same umask). So this is a little academic, but we should get it right anyway. Fixes: 47057abde515 (nfsd: add support for the umask attribute) Cc: stable@vger.kernel.org Reported-by: Lucash Stach <l.stach@pengutronix.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-03-22 05:19:02 +08:00
status = nfserr_inval;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
rdev = MKDEV(create->cr_specdata1, create->cr_specdata2);
if (MAJOR(rdev) != create->cr_specdata1 ||
MINOR(rdev) != create->cr_specdata2)
nfsd: fix incorrect umasks We're neglecting to clear the umask after it's set, which can cause a later unrelated rpc to (incorrectly) use the same umask if it happens to be processed by the same thread. There's a more subtle problem here too: An NFSv4 compound request is decoded all in one pass before any operations are executed. Currently we're setting current->fs->umask at the time we decode the compound. In theory a single compound could contain multiple creates each setting a umask. In that case we'd end up using whichever umask was passed in the *last* operation as the umask for all the creates, whether that was correct or not. So, we should just be saving the umask at decode time and waiting to set it until we actually process the corresponding operation. In practice it's unlikely any client would do multiple creates in a single compound. And even if it did they'd likely be from the same process (hence carry the same umask). So this is a little academic, but we should get it right anyway. Fixes: 47057abde515 (nfsd: add support for the umask attribute) Cc: stable@vger.kernel.org Reported-by: Lucash Stach <l.stach@pengutronix.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-03-22 05:19:02 +08:00
goto out_umask;
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
status = nfsd_create(rqstp, &cstate->current_fh,
create->cr_name, create->cr_namelen,
NFSD: introduce struct nfsd_attrs The attributes that nfsd might want to set on a file include 'struct iattr' as well as an ACL and security label. The latter two are passed around quite separately from the first, in part because they are only needed for NFSv4. This leads to some clumsiness in the code, such as the attributes NOT being set in nfsd_create_setattr(). We need to keep the directory locked until all attributes are set to ensure the file is never visibile without all its attributes. This need combined with the inconsistent handling of attributes leads to more clumsiness. As a first step towards tidying this up, introduce 'struct nfsd_attrs'. This is passed (by reference) to vfs.c functions that work with attributes, and is assembled by the various nfs*proc functions which call them. As yet only iattr is included, but future patches will expand this. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
&attrs, S_IFBLK, rdev, &resfh);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
break;
case NF4CHR:
nfsd: fix incorrect umasks We're neglecting to clear the umask after it's set, which can cause a later unrelated rpc to (incorrectly) use the same umask if it happens to be processed by the same thread. There's a more subtle problem here too: An NFSv4 compound request is decoded all in one pass before any operations are executed. Currently we're setting current->fs->umask at the time we decode the compound. In theory a single compound could contain multiple creates each setting a umask. In that case we'd end up using whichever umask was passed in the *last* operation as the umask for all the creates, whether that was correct or not. So, we should just be saving the umask at decode time and waiting to set it until we actually process the corresponding operation. In practice it's unlikely any client would do multiple creates in a single compound. And even if it did they'd likely be from the same process (hence carry the same umask). So this is a little academic, but we should get it right anyway. Fixes: 47057abde515 (nfsd: add support for the umask attribute) Cc: stable@vger.kernel.org Reported-by: Lucash Stach <l.stach@pengutronix.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-03-22 05:19:02 +08:00
status = nfserr_inval;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
rdev = MKDEV(create->cr_specdata1, create->cr_specdata2);
if (MAJOR(rdev) != create->cr_specdata1 ||
MINOR(rdev) != create->cr_specdata2)
nfsd: fix incorrect umasks We're neglecting to clear the umask after it's set, which can cause a later unrelated rpc to (incorrectly) use the same umask if it happens to be processed by the same thread. There's a more subtle problem here too: An NFSv4 compound request is decoded all in one pass before any operations are executed. Currently we're setting current->fs->umask at the time we decode the compound. In theory a single compound could contain multiple creates each setting a umask. In that case we'd end up using whichever umask was passed in the *last* operation as the umask for all the creates, whether that was correct or not. So, we should just be saving the umask at decode time and waiting to set it until we actually process the corresponding operation. In practice it's unlikely any client would do multiple creates in a single compound. And even if it did they'd likely be from the same process (hence carry the same umask). So this is a little academic, but we should get it right anyway. Fixes: 47057abde515 (nfsd: add support for the umask attribute) Cc: stable@vger.kernel.org Reported-by: Lucash Stach <l.stach@pengutronix.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-03-22 05:19:02 +08:00
goto out_umask;
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
status = nfsd_create(rqstp, &cstate->current_fh,
create->cr_name, create->cr_namelen,
NFSD: introduce struct nfsd_attrs The attributes that nfsd might want to set on a file include 'struct iattr' as well as an ACL and security label. The latter two are passed around quite separately from the first, in part because they are only needed for NFSv4. This leads to some clumsiness in the code, such as the attributes NOT being set in nfsd_create_setattr(). We need to keep the directory locked until all attributes are set to ensure the file is never visibile without all its attributes. This need combined with the inconsistent handling of attributes leads to more clumsiness. As a first step towards tidying this up, introduce 'struct nfsd_attrs'. This is passed (by reference) to vfs.c functions that work with attributes, and is assembled by the various nfs*proc functions which call them. As yet only iattr is included, but future patches will expand this. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
&attrs, S_IFCHR, rdev, &resfh);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
break;
case NF4SOCK:
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
status = nfsd_create(rqstp, &cstate->current_fh,
create->cr_name, create->cr_namelen,
NFSD: introduce struct nfsd_attrs The attributes that nfsd might want to set on a file include 'struct iattr' as well as an ACL and security label. The latter two are passed around quite separately from the first, in part because they are only needed for NFSv4. This leads to some clumsiness in the code, such as the attributes NOT being set in nfsd_create_setattr(). We need to keep the directory locked until all attributes are set to ensure the file is never visibile without all its attributes. This need combined with the inconsistent handling of attributes leads to more clumsiness. As a first step towards tidying this up, introduce 'struct nfsd_attrs'. This is passed (by reference) to vfs.c functions that work with attributes, and is assembled by the various nfs*proc functions which call them. As yet only iattr is included, but future patches will expand this. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
&attrs, S_IFSOCK, 0, &resfh);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
break;
case NF4FIFO:
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
status = nfsd_create(rqstp, &cstate->current_fh,
create->cr_name, create->cr_namelen,
NFSD: introduce struct nfsd_attrs The attributes that nfsd might want to set on a file include 'struct iattr' as well as an ACL and security label. The latter two are passed around quite separately from the first, in part because they are only needed for NFSv4. This leads to some clumsiness in the code, such as the attributes NOT being set in nfsd_create_setattr(). We need to keep the directory locked until all attributes are set to ensure the file is never visibile without all its attributes. This need combined with the inconsistent handling of attributes leads to more clumsiness. As a first step towards tidying this up, introduce 'struct nfsd_attrs'. This is passed (by reference) to vfs.c functions that work with attributes, and is assembled by the various nfs*proc functions which call them. As yet only iattr is included, but future patches will expand this. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
&attrs, S_IFIFO, 0, &resfh);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
break;
case NF4DIR:
create->cr_iattr.ia_valid &= ~ATTR_SIZE;
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
status = nfsd_create(rqstp, &cstate->current_fh,
create->cr_name, create->cr_namelen,
NFSD: introduce struct nfsd_attrs The attributes that nfsd might want to set on a file include 'struct iattr' as well as an ACL and security label. The latter two are passed around quite separately from the first, in part because they are only needed for NFSv4. This leads to some clumsiness in the code, such as the attributes NOT being set in nfsd_create_setattr(). We need to keep the directory locked until all attributes are set to ensure the file is never visibile without all its attributes. This need combined with the inconsistent handling of attributes leads to more clumsiness. As a first step towards tidying this up, introduce 'struct nfsd_attrs'. This is passed (by reference) to vfs.c functions that work with attributes, and is assembled by the various nfs*proc functions which call them. As yet only iattr is included, but future patches will expand this. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
&attrs, S_IFDIR, 0, &resfh);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
break;
default:
status = nfserr_badtype;
}
NFSv4: ACL in operations 'open' and 'create' should be used ACL in operations 'open' and 'create' is decoded but never be used. It should be set as the initial ACL for the object according to RFC3530. If error occurs when setting the ACL, just clear the ACL bit in the returned attr bitmap. Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-07-06 17:24:16 +08:00
if (status)
goto out;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
NFSD: add security label to struct nfsd_attrs nfsd_setattr() now sets a security label if provided, and nfsv4 provides it in the 'open' and 'create' paths and the 'setattr' path. If setting the label failed (including because the kernel doesn't support labels), an error field in 'struct nfsd_attrs' is set, and the caller can respond. The open/create callers clear FATTR4_WORD2_SECURITY_LABEL in the returned attr set in this case. The setattr caller returns the error. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
if (attrs.na_labelerr)
create->cr_bmval[2] &= ~FATTR4_WORD2_SECURITY_LABEL;
NFSD: add posix ACLs to struct nfsd_attrs pacl and dpacl pointers are added to struct nfsd_attrs, which requires that we have an nfsd_attrs_free() function to free them. Those nfsv4 functions that can set ACLs now set up these pointers based on the passed in NFSv4 ACL. nfsd_setattr() sets the acls as appropriate. Errors are handled as with security labels. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
if (attrs.na_aclerr)
create->cr_bmval[0] &= ~FATTR4_WORD0_ACL;
NFSv4: ACL in operations 'open' and 'create' should be used ACL in operations 'open' and 'create' is decoded but never be used. It should be set as the initial ACL for the object according to RFC3530. If error occurs when setting the ACL, just clear the ACL bit in the returned attr bitmap. Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-07-06 17:24:16 +08:00
set_change_info(&create->cr_cinfo, &cstate->current_fh);
fh_dup2(&cstate->current_fh, &resfh);
out:
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
fh_put(&resfh);
nfsd: fix incorrect umasks We're neglecting to clear the umask after it's set, which can cause a later unrelated rpc to (incorrectly) use the same umask if it happens to be processed by the same thread. There's a more subtle problem here too: An NFSv4 compound request is decoded all in one pass before any operations are executed. Currently we're setting current->fs->umask at the time we decode the compound. In theory a single compound could contain multiple creates each setting a umask. In that case we'd end up using whichever umask was passed in the *last* operation as the umask for all the creates, whether that was correct or not. So, we should just be saving the umask at decode time and waiting to set it until we actually process the corresponding operation. In practice it's unlikely any client would do multiple creates in a single compound. And even if it did they'd likely be from the same process (hence carry the same umask). So this is a little academic, but we should get it right anyway. Fixes: 47057abde515 (nfsd: add support for the umask attribute) Cc: stable@vger.kernel.org Reported-by: Lucash Stach <l.stach@pengutronix.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-03-22 05:19:02 +08:00
out_umask:
current->fs->umask = 0;
NFSD: add posix ACLs to struct nfsd_attrs pacl and dpacl pointers are added to struct nfsd_attrs, which requires that we have an nfsd_attrs_free() function to free them. Those nfsv4 functions that can set ACLs now set up these pointers based on the passed in NFSv4 ACL. nfsd_setattr() sets the acls as appropriate. Errors are handled as with security labels. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
nfsd_attrs_free(&attrs);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return status;
}
[PATCH] knfsd: nfsd4: don't inline nfsd4 compound op functions The inlining contributes to bloating the stack of nfsd4_compound, and I want to change the compound op functions to function pointers anyway. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:29 +08:00
static __be32
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
nfsd4_getattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_getattr *getattr = &u->getattr;
[PATCH] nfsd: nfs4 code returns error values in net-endian Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:28:59 +08:00
__be32 status;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
nfsd: rename MAY_ flags Rename nfsd_permission() specific MAY_* flags to NFSD_MAY_* to make it clear, that these are not used outside nfsd, and to avoid name and number space conflicts with the VFS. [comment from hch: rename MAY_READ, MAY_WRITE and MAY_EXEC as well] Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-06-16 19:20:29 +08:00
status = fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_NOP);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (status)
return status;
if (getattr->ga_bmval[1] & NFSD_WRITEONLY_ATTRS_WORD1)
return nfserr_inval;
nfsd: clean up supported attribute handling Minor cleanup, no change in behavior. Provide helpers for some common attribute bitmap operations. Drop some comments that just echo the code. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-10-19 02:18:40 +08:00
getattr->ga_bmval[0] &= nfsd_suppattrs[cstate->minorversion][0];
getattr->ga_bmval[1] &= nfsd_suppattrs[cstate->minorversion][1];
getattr->ga_bmval[2] &= nfsd_suppattrs[cstate->minorversion][2];
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
getattr->ga_fhp = &cstate->current_fh;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return nfs_ok;
}
[PATCH] knfsd: nfsd4: don't inline nfsd4 compound op functions The inlining contributes to bloating the stack of nfsd4_compound, and I want to change the compound op functions to function pointers anyway. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:29 +08:00
static __be32
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
nfsd4_link(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_link *link = &u->link;
nfsd: remove unnecessary nofilehandle checks These checks should have already be done centrally in nfsd4_proc_compound, the checks in each individual operation are unnecessary. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-09-16 04:02:52 +08:00
__be32 status;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
status = nfsd_link(rqstp, &cstate->current_fh,
link->li_name, link->li_namelen, &cstate->save_fh);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (!status)
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
set_change_info(&link->li_cinfo, &cstate->current_fh);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return status;
}
nfsd4: move guts of nfsd4_lookupp into helper We'll reuse this code in secinfo_no_name. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2010-12-16 23:06:27 +08:00
static __be32 nfsd4_do_lookupp(struct svc_rqst *rqstp, struct svc_fh *fh)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
struct svc_fh tmp_fh;
[PATCH] nfsd: nfs4 code returns error values in net-endian Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:28:59 +08:00
__be32 ret;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
fh_init(&tmp_fh, NFS4_FHSIZE);
knfsd: nfsd4: simplify exp_pseudoroot arguments We're passing three arguments to exp_pseudoroot, two of which are just fields of the svc_rqst. Soon we'll want to pass in a third field as well. So let's just give up and pass in the whole struct svc_rqst. Also sneak in some minor style cleanups while we're at it. Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-17 19:04:43 +08:00
ret = exp_pseudoroot(rqstp, &tmp_fh);
if (ret)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return ret;
nfsd4: move guts of nfsd4_lookupp into helper We'll reuse this code in secinfo_no_name. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2010-12-16 23:06:27 +08:00
if (tmp_fh.fh_dentry == fh->fh_dentry) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
fh_put(&tmp_fh);
return nfserr_noent;
}
fh_put(&tmp_fh);
nfsd4: move guts of nfsd4_lookupp into helper We'll reuse this code in secinfo_no_name. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2010-12-16 23:06:27 +08:00
return nfsd_lookup(rqstp, fh, "..", 2, fh);
}
static __be32
nfsd4_lookupp(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
nfsd4: move guts of nfsd4_lookupp into helper We'll reuse this code in secinfo_no_name. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2010-12-16 23:06:27 +08:00
{
return nfsd4_do_lookupp(rqstp, &cstate->current_fh);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
[PATCH] knfsd: nfsd4: don't inline nfsd4 compound op functions The inlining contributes to bloating the stack of nfsd4_compound, and I want to change the compound op functions to function pointers anyway. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:29 +08:00
static __be32
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
nfsd4_lookup(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
return nfsd_lookup(rqstp, &cstate->current_fh,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
u->lookup.lo_name, u->lookup.lo_len,
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
&cstate->current_fh);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
[PATCH] knfsd: nfsd4: don't inline nfsd4 compound op functions The inlining contributes to bloating the stack of nfsd4_compound, and I want to change the compound op functions to function pointers anyway. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:29 +08:00
static __be32
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
nfsd4_read(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_read *read = &u->read;
[PATCH] nfsd: nfs4 code returns error values in net-endian Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:28:59 +08:00
__be32 status;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp. Since we now presume that the struct file will be persistent in most cases, we can stop fiddling with the raparms in the read code. This also means that we don't really care about the rd_tmp_file field anymore. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-08-19 02:18:54 +08:00
read->rd_nf = NULL;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
nfsd: Add I/O trace points in the NFSv4 read proc NFSv4 read compound processing invokes nfsd_splice_read and nfs_readv directly, so the trace points currently in nfsd_read are not invoked for NFSv4 reads. Move the NFSD READ trace points to common helpers so that NFSv4 reads are captured. Also, record any local I/O error that occurs, the total count of bytes that were actually returned, and whether splice or vectored read was used. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-03-29 01:29:11 +08:00
trace_nfsd_read_start(rqstp, &cstate->current_fh,
read->rd_offset, read->rd_length);
NFSD: Fix the behavior of READ near OFFSET_MAX Dan Aloni reports: > Due to commit 8cfb9015280d ("NFS: Always provide aligned buffers to > the RPC read layers") on the client, a read of 0xfff is aligned up > to server rsize of 0x1000. > > As a result, in a test where the server has a file of size > 0x7fffffffffffffff, and the client tries to read from the offset > 0x7ffffffffffff000, the read causes loff_t overflow in the server > and it returns an NFS code of EINVAL to the client. The client as > a result indefinitely retries the request. The Linux NFS client does not handle NFS?ERR_INVAL, even though all NFS specifications permit servers to return that status code for a READ. Instead of NFS?ERR_INVAL, have out-of-range READ requests succeed and return a short result. Set the EOF flag in the result to prevent the client from retrying the READ request. This behavior appears to be consistent with Solaris NFS servers. Note that NFSv3 and NFSv4 use u64 offset values on the wire. These must be converted to loff_t internally before use -- an implicit type cast is not adequate for this purpose. Otherwise VFS checks against sb->s_maxbytes do not work properly. Reported-by: Dan Aloni <dan.aloni@vastdata.com> Cc: stable@vger.kernel.org Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-02-05 04:19:34 +08:00
read->rd_length = min_t(u32, read->rd_length, svc_max_payload(rqstp));
if (read->rd_offset > (u64)OFFSET_MAX)
read->rd_offset = (u64)OFFSET_MAX;
if (read->rd_offset + read->rd_length > (u64)OFFSET_MAX)
read->rd_length = (u64)OFFSET_MAX - read->rd_offset;
nfsd4: disable zero-copy on non-final read ops To ensure ordering of read data with any following operations, turn off zero copy if the read is not the final operation in the compound. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-12-05 07:03:46 +08:00
/*
* If we do a zero copy read, then a client will see read data
* that reflects the state of the file *after* performing the
* following compound.
*
* To ensure proper ordering, we therefore turn off zero copy if
* the client wants us to do more in this compound:
*/
NFSD: Modify NFSv4 to use nfsd_read_splice_ok() Avoid the use of an atomic bitop, and prepare for adding a run-time switch for using splice reads. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-11-18 06:14:40 +08:00
if (!nfsd4_last_compound_op(rqstp)) {
struct nfsd4_compoundargs *argp = rqstp->rq_argp;
argp->splice_ok = false;
}
nfsd4: disable zero-copy on non-final read ops To ensure ordering of read data with any following operations, turn off zero copy if the read is not the final operation in the compound. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-12-05 07:03:46 +08:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* check stateid */
nfsd: Pass filehandle to nfs4_preprocess_stateid_op() This will be needed so COPY can look up the saved_fh in addition to the current_fh. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 19:59:51 +08:00
status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
&read->rd_stateid, RD_STATE,
NFSD add COPY_NOTIFY operation Introducing the COPY_NOTIFY operation. Create a new unique stateid that will keep track of the copy state and the upcoming READs that will use that stateid. Each associated parent stateid has a list of copy notify stateids. A copy notify structure makes a copy of the parent stateid and a clientid and will use it to look up the parent stateid during the READ request (suggested by Trond Myklebust <trond.myklebust@hammerspace.com>). At nfs4_put_stid() time, we walk the list of the associated copy notify stateids and delete them. Laundromat thread will traverse globally stored copy notify stateid in idr and notice if any haven't been referenced in the lease period, if so, it'll remove them. Return single netaddr to advertise to the copy. Suggested-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Andy Adamson <andros@netapp.com>
2019-10-05 04:34:26 +08:00
&read->rd_nf, NULL);
NFSD: Clean up nfs4_preprocess_stateid_op() call sites Remove the lame-duck dprintk()s around nfs4_preprocess_stateid_op() call sites. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Tested-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: NeilBrown <neilb@suse.de>
2022-10-28 22:46:57 +08:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
read->rd_rqstp = rqstp;
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
read->rd_fhp = &cstate->current_fh;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return status;
}
nfsd4: define ->op_release for compound ops Run a separate ->op_release function if necessary instead of depending on the xdr encoder to do this. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-05-06 05:09:37 +08:00
static void
nfsd4_read_release(union nfsd4_op_u *u)
{
nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp. Since we now presume that the struct file will be persistent in most cases, we can stop fiddling with the raparms in the read code. This also means that we don't really care about the rd_tmp_file field anymore. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-08-19 02:18:54 +08:00
if (u->read.rd_nf)
nfsd_file_put(u->read.rd_nf);
nfsd: Add I/O trace points in the NFSv4 read proc NFSv4 read compound processing invokes nfsd_splice_read and nfs_readv directly, so the trace points currently in nfsd_read are not invoked for NFSv4 reads. Move the NFSD READ trace points to common helpers so that NFSv4 reads are captured. Also, record any local I/O error that occurs, the total count of bytes that were actually returned, and whether splice or vectored read was used. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-03-29 01:29:11 +08:00
trace_nfsd_read_done(u->read.rd_rqstp, u->read.rd_fhp,
u->read.rd_offset, u->read.rd_length);
nfsd4: define ->op_release for compound ops Run a separate ->op_release function if necessary instead of depending on the xdr encoder to do this. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-05-06 05:09:37 +08:00
}
[PATCH] knfsd: nfsd4: don't inline nfsd4 compound op functions The inlining contributes to bloating the stack of nfsd4_compound, and I want to change the compound op functions to function pointers anyway. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:29 +08:00
static __be32
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
nfsd4_readdir(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_readdir *readdir = &u->readdir;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
u64 cookie = readdir->rd_cookie;
static const nfs4_verifier zeroverf;
/* no need to check permission - this will be done in nfsd_readdir() */
if (readdir->rd_bmval[1] & NFSD_WRITEONLY_ATTRS_WORD1)
return nfserr_inval;
nfsd: clean up supported attribute handling Minor cleanup, no change in behavior. Provide helpers for some common attribute bitmap operations. Drop some comments that just echo the code. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-10-19 02:18:40 +08:00
readdir->rd_bmval[0] &= nfsd_suppattrs[cstate->minorversion][0];
readdir->rd_bmval[1] &= nfsd_suppattrs[cstate->minorversion][1];
readdir->rd_bmval[2] &= nfsd_suppattrs[cstate->minorversion][2];
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
nfsd4: Remove check for a 32-bit cookie in nfsd4_readdir() Fan Yong <yong.fan@whamcloud.com> noticed setting FMODE_32bithash wouldn't work with nfsd v4, as nfsd4_readdir() checks for 32 bit cookies. However, according to RFC 3530 cookies have a 64 bit type and cookies are also defined as u64 in 'struct nfsd4_readdir'. So remove the test for >32-bit values. Cc: stable@kernel.org Signed-off-by: Bernd Schubert <bernd.schubert@itwm.fraunhofer.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-08 23:38:08 +08:00
if ((cookie == 1) || (cookie == 2) ||
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
(cookie == 0 && memcmp(readdir->rd_verf.data, zeroverf.data, NFS4_VERIFIER_SIZE)))
return nfserr_bad_cookie;
readdir->rd_rqstp = rqstp;
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
readdir->rd_fhp = &cstate->current_fh;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return nfs_ok;
}
[PATCH] knfsd: nfsd4: don't inline nfsd4 compound op functions The inlining contributes to bloating the stack of nfsd4_compound, and I want to change the compound op functions to function pointers anyway. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:29 +08:00
static __be32
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
nfsd4_readlink(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
u->readlink.rl_rqstp = rqstp;
u->readlink.rl_fhp = &cstate->current_fh;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return nfs_ok;
}
[PATCH] knfsd: nfsd4: don't inline nfsd4 compound op functions The inlining contributes to bloating the stack of nfsd4_compound, and I want to change the compound op functions to function pointers anyway. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:29 +08:00
static __be32
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
nfsd4_remove(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_remove *remove = &u->remove;
[PATCH] nfsd: nfs4 code returns error values in net-endian Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:28:59 +08:00
__be32 status;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
lockd: NLM grace period shouldn't block NFSv4 opens NLM locks don't conflict with NFSv4 share reservations, so we're not going to learn anything new by watiting for them. They do conflict with NFSv4 locks and with delegations. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2015-08-07 00:47:02 +08:00
if (opens_in_grace(SVC_NET(rqstp)))
[PATCH] nfsd4: block metadata ops during grace period We currently return err_grace if a user attempts a non-reclaim open during the grace period. But we also need to prevent renames and removes, at least, to ensure clients have the chance to recover state on files before they are moved or deleted. Of course, local users could also do renames and removes during the lease period, and there's not much we can do about that. This at least will help with remote users. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-06-24 13:03:00 +08:00
return nfserr_grace;
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
status = nfsd_unlink(rqstp, &cstate->current_fh, 0,
remove->rm_name, remove->rm_namelen);
NFSD: always drop directory lock in nfsd_unlink() Some error paths in nfsd_unlink() allow it to exit without unlocking the directory. This is not a problem in practice as the directory will be locked with an fh_put(), but it is untidy and potentially confusing. This allows us to remove all the fh_unlock() calls that are immediately after nfsd_unlink() calls. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
if (!status)
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
set_change_info(&remove->rm_cinfo, &cstate->current_fh);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return status;
}
[PATCH] knfsd: nfsd4: don't inline nfsd4 compound op functions The inlining contributes to bloating the stack of nfsd4_compound, and I want to change the compound op functions to function pointers anyway. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:29 +08:00
static __be32
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
nfsd4_rename(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_rename *rename = &u->rename;
nfsd: remove unnecessary nofilehandle checks These checks should have already be done centrally in nfsd4_proc_compound, the checks in each individual operation are unnecessary. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-09-16 04:02:52 +08:00
__be32 status;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
nfsd4: forbid all renames during grace period The idea here was that renaming a file on a nosubtreecheck export would make lookups of the old filehandle return STALE, making it impossible for clients to reclaim opens. But during the grace period I think we should also hold off on operations that would break delegations. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-11-07 00:47:20 +08:00
if (opens_in_grace(SVC_NET(rqstp)))
[PATCH] nfsd4: block metadata ops during grace period We currently return err_grace if a user attempts a non-reclaim open during the grace period. But we also need to prevent renames and removes, at least, to ensure clients have the chance to recover state on files before they are moved or deleted. Of course, local users could also do renames and removes during the lease period, and there's not much we can do about that. This at least will help with remote users. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-06-24 13:03:00 +08:00
return nfserr_grace;
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
status = nfsd_rename(rqstp, &cstate->save_fh, rename->rn_sname,
rename->rn_snamelen, &cstate->current_fh,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
rename->rn_tname, rename->rn_tnamelen);
nfsd4: don't remap EISDIR errors in rename We're going out of our way here to remap an error to make rfc 3530 happy--but the rfc itself (nor rfc 1813, which has similar language) gives no justification. And disagrees with local filesystem behavior, with Linux and posix man pages, and knfsd's implemented behavior for v2 and v3. And the documented behavior seems better, in that it gives a little more information--you could implement the 3530 behavior using the posix behavior, but not the other way around. Also, the Linux client makes no attempt to remap this error in the v4 case, so it can end up just returning EEXIST to the application in a case where it should return EISDIR. So honestly I think the rfc's are just buggy here--or in any case it doesn't see worth the trouble to remap this error. Reported-by: Frank S Filz <ffilz@us.ibm.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-05-01 03:28:51 +08:00
if (status)
return status;
nfsd: fix change_info in NFSv4 RENAME replies nfsd sends the transposed directory change info in the RENAME reply. The source directory is in save_fh and the target is in current_fh. Reported-by: Zhi Li <yieli@redhat.com> Reported-by: Benjamin Coddington <bcodding@redhat.com> Closes: https://bugzilla.redhat.com/show_bug.cgi?id=2218844 Signed-off-by: Jeff Layton <jlayton@kernel.org> Cc: <stable@vger.kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-09-09 19:12:30 +08:00
set_change_info(&rename->rn_sinfo, &cstate->save_fh);
set_change_info(&rename->rn_tinfo, &cstate->current_fh);
nfsd4: don't remap EISDIR errors in rename We're going out of our way here to remap an error to make rfc 3530 happy--but the rfc itself (nor rfc 1813, which has similar language) gives no justification. And disagrees with local filesystem behavior, with Linux and posix man pages, and knfsd's implemented behavior for v2 and v3. And the documented behavior seems better, in that it gives a little more information--you could implement the 3530 behavior using the posix behavior, but not the other way around. Also, the Linux client makes no attempt to remap this error in the v4 case, so it can end up just returning EEXIST to the application in a case where it should return EISDIR. So honestly I think the rfc's are just buggy here--or in any case it doesn't see worth the trouble to remap this error. Reported-by: Frank S Filz <ffilz@us.ibm.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-05-01 03:28:51 +08:00
return nfs_ok;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
knfsd: nfsd4: implement secinfo Implement the secinfo operation. (Thanks to Usha Ketineni wrote an earlier version of this support.) Cc: Usha Ketineni <uketinen@us.ibm.com> Signed-off-by: Andy Adamson <andros@citi.umich.edu> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-17 19:04:51 +08:00
static __be32
nfsd4_secinfo(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
knfsd: nfsd4: implement secinfo Implement the secinfo operation. (Thanks to Usha Ketineni wrote an earlier version of this support.) Cc: Usha Ketineni <uketinen@us.ibm.com> Signed-off-by: Andy Adamson <andros@citi.umich.edu> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-17 19:04:51 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_secinfo *secinfo = &u->secinfo;
knfsd: nfsd4: implement secinfo Implement the secinfo operation. (Thanks to Usha Ketineni wrote an earlier version of this support.) Cc: Usha Ketineni <uketinen@us.ibm.com> Signed-off-by: Andy Adamson <andros@citi.umich.edu> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-17 19:04:51 +08:00
struct svc_export *exp;
struct dentry *dentry;
__be32 err;
nfsd4: make fh_verify responsibility of nfsd_lookup_dentry caller The secinfo caller actually won't want this. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-04-09 23:28:53 +08:00
err = fh_verify(rqstp, &cstate->current_fh, S_IFDIR, NFSD_MAY_EXEC);
if (err)
return err;
knfsd: nfsd4: implement secinfo Implement the secinfo operation. (Thanks to Usha Ketineni wrote an earlier version of this support.) Cc: Usha Ketineni <uketinen@us.ibm.com> Signed-off-by: Andy Adamson <andros@citi.umich.edu> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-17 19:04:51 +08:00
err = nfsd_lookup_dentry(rqstp, &cstate->current_fh,
secinfo->si_name, secinfo->si_namelen,
&exp, &dentry);
if (err)
return err;
VFS: normal filesystems (and lustre): d_inode() annotations that's the bulk of filesystem drivers dealing with inodes of their own Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-03-18 06:25:59 +08:00
if (d_really_is_negative(dentry)) {
knfsd: nfsd4: implement secinfo Implement the secinfo operation. (Thanks to Usha Ketineni wrote an earlier version of this support.) Cc: Usha Ketineni <uketinen@us.ibm.com> Signed-off-by: Andy Adamson <andros@citi.umich.edu> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-17 19:04:51 +08:00
exp_put(exp);
err = nfserr_noent;
} else
secinfo->si_exp = exp;
dput(dentry);
nfsd4: 4.1 SECINFO should consume filehandle See the referenced spec language; an attempt by a 4.1 client to use the current filehandle after a secinfo call should result in a NOFILEHANDLE error. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2010-12-16 22:57:15 +08:00
if (cstate->minorversion)
/* See rfc 5661 section 2.6.3.1.1.8 */
fh_put(&cstate->current_fh);
knfsd: nfsd4: implement secinfo Implement the secinfo operation. (Thanks to Usha Ketineni wrote an earlier version of this support.) Cc: Usha Ketineni <uketinen@us.ibm.com> Signed-off-by: Andy Adamson <andros@citi.umich.edu> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-17 19:04:51 +08:00
return err;
}
nfsd4: implement secinfo_no_name Implementation of this operation is mandatory for NFSv4.1. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2010-12-16 22:51:13 +08:00
static __be32
nfsd4_secinfo_no_name(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
nfsd4: implement secinfo_no_name Implementation of this operation is mandatory for NFSv4.1. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2010-12-16 22:51:13 +08:00
{
__be32 err;
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
switch (u->secinfo_no_name.sin_style) {
nfsd4: implement secinfo_no_name Implementation of this operation is mandatory for NFSv4.1. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2010-12-16 22:51:13 +08:00
case NFS4_SECINFO_STYLE4_CURRENT_FH:
break;
case NFS4_SECINFO_STYLE4_PARENT:
err = nfsd4_do_lookupp(rqstp, &cstate->current_fh);
if (err)
return err;
break;
default:
return nfserr_inval;
}
NFSD: Using exp_get for export getting Don't using cache_get besides export.h, using exp_get for export. Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-06-10 22:06:44 +08:00
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
u->secinfo_no_name.sin_exp = exp_get(cstate->current_fh.fh_export);
nfsd4: implement secinfo_no_name Implementation of this operation is mandatory for NFSv4.1. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2010-12-16 22:51:13 +08:00
fh_put(&cstate->current_fh);
return nfs_ok;
}
nfsd4: define ->op_release for compound ops Run a separate ->op_release function if necessary instead of depending on the xdr encoder to do this. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-05-06 05:09:37 +08:00
static void
nfsd4_secinfo_release(union nfsd4_op_u *u)
{
if (u->secinfo.si_exp)
exp_put(u->secinfo.si_exp);
}
nfsd4: define nfsd4_secinfo_no_name_release() Commit 34b1744c91cc ("nfsd4: define ->op_release for compound ops") defined a couple ->op_release functions and run them if necessary. But there's a problem with that is that it reused nfsd4_secinfo_release() as the op_release of OP_SECINFO_NO_NAME, and caused a leak on struct nfsd4_secinfo_no_name in nfsd4_encode_secinfo_no_name(), because there's no .si_exp field in struct nfsd4_secinfo_no_name. I found this because I was unable to umount an ext4 partition after exporting it via NFS & run fsstress on the nfs mount. A simplified reproducer would be: # mount a local-fs device at /mnt/test, and export it via NFS with # fsid=0 export option (this is required) mount /dev/sda5 /mnt/test echo "/mnt/test *(rw,no_root_squash,fsid=0)" >> /etc/exports service nfs restart # locally mount the nfs export with all default, note that I have # nfsv4.1 configured as the default nfs version, because of the # fsid export option, v4 mount would fail and fall back to v3 mount localhost:/mnt/test /mnt/nfs # try to umount the underlying device, but got EBUSY umount /mnt/nfs service nfs stop umount /mnt/test <=== EBUSY here Fixed it by defining a separate nfsd4_secinfo_no_name_release() function as the op_release method of OP_SECINFO_NO_NAME that releases the correct nfsd4_secinfo_no_name structure. Fixes: 34b1744c91cc ("nfsd4: define ->op_release for compound ops") Signed-off-by: Eryu Guan <eguan@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-09-29 15:01:10 +08:00
static void
nfsd4_secinfo_no_name_release(union nfsd4_op_u *u)
{
if (u->secinfo_no_name.sin_exp)
exp_put(u->secinfo_no_name.sin_exp);
}
[PATCH] knfsd: nfsd4: don't inline nfsd4 compound op functions The inlining contributes to bloating the stack of nfsd4_compound, and I want to change the compound op functions to function pointers anyway. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:29 +08:00
static __be32
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
nfsd4_setattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_setattr *setattr = &u->setattr;
NFSD: introduce struct nfsd_attrs The attributes that nfsd might want to set on a file include 'struct iattr' as well as an ACL and security label. The latter two are passed around quite separately from the first, in part because they are only needed for NFSv4. This leads to some clumsiness in the code, such as the attributes NOT being set in nfsd_create_setattr(). We need to keep the directory locked until all attributes are set to ensure the file is never visibile without all its attributes. This need combined with the inconsistent handling of attributes leads to more clumsiness. As a first step towards tidying this up, introduce 'struct nfsd_attrs'. This is passed (by reference) to vfs.c functions that work with attributes, and is assembled by the various nfs*proc functions which call them. As yet only iattr is included, but future patches will expand this. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
struct nfsd_attrs attrs = {
.na_iattr = &setattr->sa_iattr,
NFSD: add security label to struct nfsd_attrs nfsd_setattr() now sets a security label if provided, and nfsv4 provides it in the 'open' and 'create' paths and the 'setattr' path. If setting the label failed (including because the kernel doesn't support labels), an error field in 'struct nfsd_attrs' is set, and the caller can respond. The open/create callers clear FATTR4_WORD2_SECURITY_LABEL in the returned attr set in this case. The setattr caller returns the error. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
.na_seclabel = &setattr->sa_label,
NFSD: introduce struct nfsd_attrs The attributes that nfsd might want to set on a file include 'struct iattr' as well as an ACL and security label. The latter two are passed around quite separately from the first, in part because they are only needed for NFSv4. This leads to some clumsiness in the code, such as the attributes NOT being set in nfsd_create_setattr(). We need to keep the directory locked until all attributes are set to ensure the file is never visibile without all its attributes. This need combined with the inconsistent handling of attributes leads to more clumsiness. As a first step towards tidying this up, introduce 'struct nfsd_attrs'. This is passed (by reference) to vfs.c functions that work with attributes, and is assembled by the various nfs*proc functions which call them. As yet only iattr is included, but future patches will expand this. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
};
NFSD: add posix ACLs to struct nfsd_attrs pacl and dpacl pointers are added to struct nfsd_attrs, which requires that we have an nfsd_attrs_free() function to free them. Those nfsv4 functions that can set ACLs now set up these pointers based on the passed in NFSv4 ACL. nfsd_setattr() sets the acls as appropriate. Errors are handled as with security labels. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
struct inode *inode;
[PATCH] nfsd: nfs4 code returns error values in net-endian Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:28:59 +08:00
__be32 status = nfs_ok;
nfsd: Fix a regression in nfsd_setattr() Commit bb4d53d66e4b ("NFSD: use (un)lock_inode instead of fh_(un)lock for file operations") broke the NFSv3 pre/post op attributes behaviour when doing a SETATTR rpc call by stripping out the calls to fh_fill_pre_attrs() and fh_fill_post_attrs(). Fixes: bb4d53d66e4b ("NFSD: use (un)lock_inode instead of fh_(un)lock for file operations") Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: NeilBrown <neilb@suse.de> Message-ID: <20240216012451.22725-1-trondmy@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-02-16 09:24:50 +08:00
bool save_no_wcc;
nfsd: fix b0rken error value for setattr on read-only mount ..._want_write() returns -EROFS on failure, _not_ an NFS error value. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-04-13 11:47:00 +08:00
int err;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (setattr->sa_iattr.ia_valid & ATTR_SIZE) {
nfsd: take struct file setup fully into nfs4_preprocess_stateid_op This patch changes nfs4_preprocess_stateid_op so it always returns a valid struct file if it has been asked for that. For that we now allocate a temporary struct file for special stateids, and check permissions if we got the file structure from the stateid. This ensures that all callers will get their handling of special stateids right, and avoids code duplication. There is a little wart in here because the read code needs to know if we allocated a file structure so that it can copy around the read-ahead parameters. In the long run we should probably aim to cache full file structures used with special stateids instead. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2015-06-18 22:45:00 +08:00
status = nfs4_preprocess_stateid_op(rqstp, cstate,
nfsd: Pass filehandle to nfs4_preprocess_stateid_op() This will be needed so COPY can look up the saved_fh in addition to the current_fh. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 19:59:51 +08:00
&cstate->current_fh, &setattr->sa_stateid,
NFSD add COPY_NOTIFY operation Introducing the COPY_NOTIFY operation. Create a new unique stateid that will keep track of the copy state and the upcoming READs that will use that stateid. Each associated parent stateid has a list of copy notify stateids. A copy notify structure makes a copy of the parent stateid and a clientid and will use it to look up the parent stateid during the READ request (suggested by Trond Myklebust <trond.myklebust@hammerspace.com>). At nfs4_put_stid() time, we walk the list of the associated copy notify stateids and delete them. Laundromat thread will traverse globally stored copy notify stateid in idr and notice if any haven't been referenced in the lease period, if so, it'll remove them. Return single netaddr to advertise to the copy. Suggested-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Andy Adamson <andros@netapp.com>
2019-10-05 04:34:26 +08:00
WR_STATE, NULL, NULL);
NFSD: Clean up nfs4_preprocess_stateid_op() call sites Remove the lame-duck dprintk()s around nfs4_preprocess_stateid_op() call sites. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Tested-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: NeilBrown <neilb@suse.de>
2022-10-28 22:46:57 +08:00
if (status)
[PATCH] nfsd4: nfs4state.c miscellaneous goto removals Remove some goto's that made the logic here a little more tortuous than necessary. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-19 09:43:33 +08:00
return status;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
nfsd: fix b0rken error value for setattr on read-only mount ..._want_write() returns -EROFS on failure, _not_ an NFS error value. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-04-13 11:47:00 +08:00
err = fh_want_write(&cstate->current_fh);
if (err)
return nfserrno(err);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
status = nfs_ok;
NFSv4: do exact check about attribute specified Server should return NFS4ERR_ATTRNOTSUPP if an attribute specified is not supported in current environment. Operations CREATE, NVERIFY, OPEN, SETATTR and VERIFY should do this check. This bug is found when do newpynfs tests. The names of the tests that failed are following: CR12 NVF7a NVF7b NVF7c NVF7d NVF7f NVF7r NVF7s OPEN15 VF7a VF7b VF7c VF7d VF7f VF7r VF7s Add function do_check_fattr() to do exact check: 1, Check attribute specified is supported by the NFSv4 server or not. 2, Check FATTR4_WORD0_ACL & FATTR4_WORD0_FS_LOCATIONS are supported in current environment or not. 3, Check attribute specified is writable or not. step 1 and 3 are done in function nfsd4_decode_fattr() but removed to this function now. Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-05-16 16:22:31 +08:00
status = check_attr_support(rqstp, cstate, setattr->sa_bmval,
nfsd_attrmask);
if (status)
goto out;
NFSD: add posix ACLs to struct nfsd_attrs pacl and dpacl pointers are added to struct nfsd_attrs, which requires that we have an nfsd_attrs_free() function to free them. Those nfsv4 functions that can set ACLs now set up these pointers based on the passed in NFSv4 ACL. nfsd_setattr() sets the acls as appropriate. Errors are handled as with security labels. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
inode = cstate->current_fh.fh_dentry->d_inode;
status = nfsd4_acl_to_attr(S_ISDIR(inode->i_mode) ? NF4DIR : NF4REG,
setattr->sa_acl, &attrs);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (status)
[PATCH] r/o bind mounts: elevate write count for xattr_permission() callers This basically audits the callers of xattr_permission(), which calls permission() and can perform writes to the filesystem. [AV: add missing parts - removexattr() and nfsd posix acls, plug for a leak spotted by Miklos] Acked-by: Al Viro <viro@ZenIV.linux.org.uk> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Dave Hansen <haveblue@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2008-02-16 06:37:38 +08:00
goto out;
nfsd: Fix a regression in nfsd_setattr() Commit bb4d53d66e4b ("NFSD: use (un)lock_inode instead of fh_(un)lock for file operations") broke the NFSv3 pre/post op attributes behaviour when doing a SETATTR rpc call by stripping out the calls to fh_fill_pre_attrs() and fh_fill_post_attrs(). Fixes: bb4d53d66e4b ("NFSD: use (un)lock_inode instead of fh_(un)lock for file operations") Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: NeilBrown <neilb@suse.de> Message-ID: <20240216012451.22725-1-trondmy@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-02-16 09:24:50 +08:00
save_no_wcc = cstate->current_fh.fh_no_wcc;
cstate->current_fh.fh_no_wcc = true;
nfsd: Fix NFSv3 atomicity bugs in nfsd_setattr() The main point of the guarded SETATTR is to prevent races with other WRITE and SETATTR calls. That requires that the check of the guard time against the inode ctime be done after taking the inode lock. Furthermore, we need to take into account the 32-bit nature of timestamps in NFSv3, and the possibility that files may change at a faster rate than once a second. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-02-16 09:24:51 +08:00
status = nfsd_setattr(rqstp, &cstate->current_fh, &attrs, NULL);
nfsd: Fix a regression in nfsd_setattr() Commit bb4d53d66e4b ("NFSD: use (un)lock_inode instead of fh_(un)lock for file operations") broke the NFSv3 pre/post op attributes behaviour when doing a SETATTR rpc call by stripping out the calls to fh_fill_pre_attrs() and fh_fill_post_attrs(). Fixes: bb4d53d66e4b ("NFSD: use (un)lock_inode instead of fh_(un)lock for file operations") Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: NeilBrown <neilb@suse.de> Message-ID: <20240216012451.22725-1-trondmy@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-02-16 09:24:50 +08:00
cstate->current_fh.fh_no_wcc = save_no_wcc;
NFSD: add security label to struct nfsd_attrs nfsd_setattr() now sets a security label if provided, and nfsv4 provides it in the 'open' and 'create' paths and the 'setattr' path. If setting the label failed (including because the kernel doesn't support labels), an error field in 'struct nfsd_attrs' is set, and the caller can respond. The open/create callers clear FATTR4_WORD2_SECURITY_LABEL in the returned attr set in this case. The setattr caller returns the error. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
if (!status)
status = nfserrno(attrs.na_labelerr);
nfsd: return error if nfs4_setacl fails With the addition of POSIX ACLs to struct nfsd_attrs, we no longer return an error if setting the ACL fails. Ensure we return the na_aclerr error on SETATTR if there is one. Fixes: c0cbe70742f4 ("NFSD: add posix ACLs to struct nfsd_attrs") Cc: Neil Brown <neilb@suse.de> Reported-by: Yongcheng Yang <yoyang@redhat.com> Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-11-07 19:58:41 +08:00
if (!status)
status = nfserrno(attrs.na_aclerr);
[PATCH] r/o bind mounts: elevate write count for xattr_permission() callers This basically audits the callers of xattr_permission(), which calls permission() and can perform writes to the filesystem. [AV: add missing parts - removexattr() and nfsd posix acls, plug for a leak spotted by Miklos] Acked-by: Al Viro <viro@ZenIV.linux.org.uk> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Dave Hansen <haveblue@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2008-02-16 06:37:38 +08:00
out:
NFSD: add posix ACLs to struct nfsd_attrs pacl and dpacl pointers are added to struct nfsd_attrs, which requires that we have an nfsd_attrs_free() function to free them. Those nfsv4 functions that can set ACLs now set up these pointers based on the passed in NFSv4 ACL. nfsd_setattr() sets the acls as appropriate. Errors are handled as with security labels. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-26 14:45:30 +08:00
nfsd_attrs_free(&attrs);
new helpers: fh_{want,drop}_write() A bunch of places in nfsd does mnt_{want,drop}_write on vfsmount of export of given fhandle. Switched to obvious inlined helpers... Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2011-11-24 01:03:18 +08:00
fh_drop_write(&cstate->current_fh);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return status;
}
[PATCH] knfsd: nfsd4: don't inline nfsd4 compound op functions The inlining contributes to bloating the stack of nfsd4_compound, and I want to change the compound op functions to function pointers anyway. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:29 +08:00
static __be32
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
nfsd4_write(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_write *write = &u->write;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
stateid_t *stateid = &write->wr_stateid;
nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp. Since we now presume that the struct file will be persistent in most cases, we can stop fiddling with the raparms in the read code. This also means that we don't really care about the rd_tmp_file field anymore. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-08-19 02:18:54 +08:00
struct nfsd_file *nf = NULL;
[PATCH] nfsd: nfs4 code returns error values in net-endian Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:28:59 +08:00
__be32 status = nfs_ok;
Short write in nfsd becomes a full write to the client If a filesystem being written to via NFS returns a short write count (as opposed to an error) to nfsd, nfsd treats that as a success for the entire write, rather than the short count that actually succeeded. For example, given a 8192 byte write, if the underlying filesystem only writes 4096 bytes, nfsd will ack back to the nfs client that all 8192 bytes were written. The nfs client does have retry logic for short writes, but this is never called as the client is told the complete write succeeded. There are probably other ways it could happen, but in my case it happened with a fuse (filesystem in userspace) filesystem which can rather easily have a partial write. Here is a patch to properly return the short write count to the client. Signed-off-by: David Shaw <dshaw@jabberwocky.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-03-06 09:16:14 +08:00
unsigned long cnt;
nfsd4: delay filling in write iovec array till after xdr decoding Our server rejects compounds containing more than one write operation. It's unclear whether this is really permitted by the spec; with 4.0, it's possibly OK, with 4.1 (which has clearer limits on compound parameters), it's probably not OK. No client that we're aware of has ever done this, but in theory it could be useful. The source of the limitation: we need an array of iovecs to pass to the write operation. In the worst case that array of iovecs could have hundreds of elements (the maximum rwsize divided by the page size), so it's too big to put on the stack, or in each compound op. So we instead keep a single such array in the compound argument. We fill in that array at the time we decode the xdr operation. But we decode every op in the compound before executing any of them. So once we've used that array we can't decode another write. If we instead delay filling in that array till the time we actually perform the write, we can reuse it. Another option might be to switch to decoding compound ops one at a time. I considered doing that, but it has a number of other side effects, and I'd rather fix just this one problem for now. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-11-16 03:52:19 +08:00
int nvecs;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
NFSD: Clamp WRITE offsets Ensure that a client cannot specify a WRITE range that falls in a byte range outside what the kernel's internal types (such as loff_t, which is signed) can represent. The kiocb iterators, invoked in nfsd_vfs_write(), should properly limit write operations to within the underlying file system's s_maxbytes. Cc: stable@vger.kernel.org Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-01-26 05:36:22 +08:00
if (write->wr_offset > (u64)OFFSET_MAX ||
write->wr_offset + write->wr_buflen > (u64)OFFSET_MAX)
return nfserr_fbig;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
nfsd: Add I/O trace points in the NFSv4 write path NFSv4 write compound processing invokes nfsd_vfs_write directly. The trace points currently in nfsd_write are not effective for NFSv4 writes. Move the trace points into the shared nfsd_vfs_write() helper. After the I/O, we also want to record any local I/O error that might have occurred, and the total count of bytes that were actually moved (rather than the requested number). Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-03-27 22:53:27 +08:00
cnt = write->wr_buflen;
trace_nfsd_write_start(rqstp, &cstate->current_fh,
write->wr_offset, cnt);
nfsd: Pass filehandle to nfs4_preprocess_stateid_op() This will be needed so COPY can look up the saved_fh in addition to the current_fh. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 19:59:51 +08:00
status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
NFSD add COPY_NOTIFY operation Introducing the COPY_NOTIFY operation. Create a new unique stateid that will keep track of the copy state and the upcoming READs that will use that stateid. Each associated parent stateid has a list of copy notify stateids. A copy notify structure makes a copy of the parent stateid and a clientid and will use it to look up the parent stateid during the READ request (suggested by Trond Myklebust <trond.myklebust@hammerspace.com>). At nfs4_put_stid() time, we walk the list of the associated copy notify stateids and delete them. Laundromat thread will traverse globally stored copy notify stateid in idr and notice if any haven't been referenced in the lease period, if so, it'll remove them. Return single netaddr to advertise to the copy. Suggested-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Andy Adamson <andros@netapp.com>
2019-10-05 04:34:26 +08:00
stateid, WR_STATE, &nf, NULL);
NFSD: Clean up nfs4_preprocess_stateid_op() call sites Remove the lame-duck dprintk()s around nfs4_preprocess_stateid_op() call sites. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Tested-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: NeilBrown <neilb@suse.de>
2022-10-28 22:46:57 +08:00
if (status)
[PATCH] nfsd4: nfs4state.c miscellaneous goto removals Remove some goto's that made the logic here a little more tortuous than necessary. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-19 09:43:33 +08:00
return status;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
write->wr_how_written = write->wr_stable_how;
NFSD: Have legacy NFSD WRITE decoders use xdr_stream_subsegment() Refactor. Now that the NFSv2 and NFSv3 XDR decoders have been converted to use xdr_streams, the WRITE decoder functions can use xdr_stream_subsegment() to extract the WRITE payload into its own xdr_buf, just as the NFSv4 WRITE XDR decoder currently does. That makes it possible to pass the first kvec, pages array + length, page_base, and total payload length via a single function parameter. The payload's page_base is not yet assigned or used, but will be in subsequent patches. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-10-01 05:06:21 +08:00
nvecs = svc_fill_write_vector(rqstp, &write->wr_payload);
nfsd4: delay filling in write iovec array till after xdr decoding Our server rejects compounds containing more than one write operation. It's unclear whether this is really permitted by the spec; with 4.0, it's possibly OK, with 4.1 (which has clearer limits on compound parameters), it's probably not OK. No client that we're aware of has ever done this, but in theory it could be useful. The source of the limitation: we need an array of iovecs to pass to the write operation. In the worst case that array of iovecs could have hundreds of elements (the maximum rwsize divided by the page size), so it's too big to put on the stack, or in each compound op. So we instead keep a single such array in the compound argument. We fill in that array at the time we decode the xdr operation. But we decode every op in the compound before executing any of them. So once we've used that array we can't decode another write. If we instead delay filling in that array till the time we actually perform the write, we can reuse it. Another option might be to switch to decoding compound ops one at a time. I considered doing that, but it has a number of other side effects, and I'd rather fix just this one problem for now. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-11-16 03:52:19 +08:00
WARN_ON_ONCE(nvecs > ARRAY_SIZE(rqstp->rq_vec));
nfsd: Allow nfsd_vfs_write() to take the nfsd_file as an argument Needed in order to fix stable writes. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2020-01-07 02:40:29 +08:00
status = nfsd_vfs_write(rqstp, &cstate->current_fh, nf,
nfsd: take struct file setup fully into nfs4_preprocess_stateid_op This patch changes nfs4_preprocess_stateid_op so it always returns a valid struct file if it has been asked for that. For that we now allocate a temporary struct file for special stateids, and check permissions if we got the file structure from the stateid. This ensures that all callers will get their handling of special stateids right, and avoids code duplication. There is a little wart in here because the read code needs to know if we allocated a file structure so that it can copy around the read-ahead parameters. In the long run we should probably aim to cache full file structures used with special stateids instead. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2015-06-18 22:45:00 +08:00
write->wr_offset, rqstp->rq_vec, nvecs, &cnt,
nfsd: Ensure sampling of the write verifier is atomic with the write When doing an unstable write, we need to ensure that we sample the write verifier before releasing the lock, and allowing a commit to the same file to proceed. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2020-01-07 02:40:37 +08:00
write->wr_how_written,
(__be32 *)write->wr_verifier.data);
nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp. Since we now presume that the struct file will be persistent in most cases, we can stop fiddling with the raparms in the read code. This also means that we don't really care about the rd_tmp_file field anymore. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-08-19 02:18:54 +08:00
nfsd_file_put(nf);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
Short write in nfsd becomes a full write to the client If a filesystem being written to via NFS returns a short write count (as opposed to an error) to nfsd, nfsd treats that as a success for the entire write, rather than the short count that actually succeeded. For example, given a 8192 byte write, if the underlying filesystem only writes 4096 bytes, nfsd will ack back to the nfs client that all 8192 bytes were written. The nfs client does have retry logic for short writes, but this is never called as the client is told the complete write succeeded. There are probably other ways it could happen, but in my case it happened with a fuse (filesystem in userspace) filesystem which can rather easily have a partial write. Here is a patch to properly return the short write count to the client. Signed-off-by: David Shaw <dshaw@jabberwocky.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-03-06 09:16:14 +08:00
write->wr_bytes_written = cnt;
nfsd: Add I/O trace points in the NFSv4 write path NFSv4 write compound processing invokes nfsd_vfs_write directly. The trace points currently in nfsd_write are not effective for NFSv4 writes. Move the trace points into the shared nfsd_vfs_write() helper. After the I/O, we also want to record any local I/O error that might have occurred, and the total count of bytes that were actually moved (rather than the requested number). Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-03-27 22:53:27 +08:00
trace_nfsd_write_done(rqstp, &cstate->current_fh,
write->wr_offset, cnt);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return status;
}
nfsd: implement the NFSv4.2 CLONE operation This is basically a remote version of the btrfs CLONE operation, so the implementation is fairly trivial. Made even more trivial by stealing the XDR code and general framework Anna Schumaker's COPY prototype. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 19:59:52 +08:00
static __be32
NFSD: Implement the COPY call I only implemented the sync version of this call, since it's the easiest. I can simply call vfs_copy_range() and have the vfs do the right thing for the filesystem being exported. Signed-off-by: Anna Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-09-08 03:57:30 +08:00
nfsd4_verify_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp. Since we now presume that the struct file will be persistent in most cases, we can stop fiddling with the raparms in the read code. This also means that we don't really care about the rd_tmp_file field anymore. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-08-19 02:18:54 +08:00
stateid_t *src_stateid, struct nfsd_file **src,
stateid_t *dst_stateid, struct nfsd_file **dst)
nfsd: implement the NFSv4.2 CLONE operation This is basically a remote version of the btrfs CLONE operation, so the implementation is fairly trivial. Made even more trivial by stealing the XDR code and general framework Anna Schumaker's COPY prototype. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 19:59:52 +08:00
{
__be32 status;
nfsd: COPY and CLONE operations require the saved filehandle to be set Make sure we have a saved filehandle, otherwise we'll oops with a null pointer dereference in nfs4_preprocess_stateid_op(). Signed-off-by: Scott Mayhew <smayhew@redhat.com> Cc: stable@vger.kernel.org Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-11-09 00:11:36 +08:00
if (!cstate->save_fh.fh_dentry)
return nfserr_nofilehandle;
nfsd: implement the NFSv4.2 CLONE operation This is basically a remote version of the btrfs CLONE operation, so the implementation is fairly trivial. Made even more trivial by stealing the XDR code and general framework Anna Schumaker's COPY prototype. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 19:59:52 +08:00
status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->save_fh,
NFSD add COPY_NOTIFY operation Introducing the COPY_NOTIFY operation. Create a new unique stateid that will keep track of the copy state and the upcoming READs that will use that stateid. Each associated parent stateid has a list of copy notify stateids. A copy notify structure makes a copy of the parent stateid and a clientid and will use it to look up the parent stateid during the READ request (suggested by Trond Myklebust <trond.myklebust@hammerspace.com>). At nfs4_put_stid() time, we walk the list of the associated copy notify stateids and delete them. Laundromat thread will traverse globally stored copy notify stateid in idr and notice if any haven't been referenced in the lease period, if so, it'll remove them. Return single netaddr to advertise to the copy. Suggested-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Andy Adamson <andros@netapp.com>
2019-10-05 04:34:26 +08:00
src_stateid, RD_STATE, src, NULL);
NFSD: Clean up nfs4_preprocess_stateid_op() call sites Remove the lame-duck dprintk()s around nfs4_preprocess_stateid_op() call sites. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Tested-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: NeilBrown <neilb@suse.de>
2022-10-28 22:46:57 +08:00
if (status)
nfsd: implement the NFSv4.2 CLONE operation This is basically a remote version of the btrfs CLONE operation, so the implementation is fairly trivial. Made even more trivial by stealing the XDR code and general framework Anna Schumaker's COPY prototype. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 19:59:52 +08:00
goto out;
status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
NFSD add COPY_NOTIFY operation Introducing the COPY_NOTIFY operation. Create a new unique stateid that will keep track of the copy state and the upcoming READs that will use that stateid. Each associated parent stateid has a list of copy notify stateids. A copy notify structure makes a copy of the parent stateid and a clientid and will use it to look up the parent stateid during the READ request (suggested by Trond Myklebust <trond.myklebust@hammerspace.com>). At nfs4_put_stid() time, we walk the list of the associated copy notify stateids and delete them. Laundromat thread will traverse globally stored copy notify stateid in idr and notice if any haven't been referenced in the lease period, if so, it'll remove them. Return single netaddr to advertise to the copy. Suggested-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Andy Adamson <andros@netapp.com>
2019-10-05 04:34:26 +08:00
dst_stateid, WR_STATE, dst, NULL);
NFSD: Clean up nfs4_preprocess_stateid_op() call sites Remove the lame-duck dprintk()s around nfs4_preprocess_stateid_op() call sites. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Tested-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: NeilBrown <neilb@suse.de>
2022-10-28 22:46:57 +08:00
if (status)
nfsd: implement the NFSv4.2 CLONE operation This is basically a remote version of the btrfs CLONE operation, so the implementation is fairly trivial. Made even more trivial by stealing the XDR code and general framework Anna Schumaker's COPY prototype. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 19:59:52 +08:00
goto out_put_src;
/* fix up for NFS-specific error code */
nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp. Since we now presume that the struct file will be persistent in most cases, we can stop fiddling with the raparms in the read code. This also means that we don't really care about the rd_tmp_file field anymore. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-08-19 02:18:54 +08:00
if (!S_ISREG(file_inode((*src)->nf_file)->i_mode) ||
!S_ISREG(file_inode((*dst)->nf_file)->i_mode)) {
nfsd: implement the NFSv4.2 CLONE operation This is basically a remote version of the btrfs CLONE operation, so the implementation is fairly trivial. Made even more trivial by stealing the XDR code and general framework Anna Schumaker's COPY prototype. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 19:59:52 +08:00
status = nfserr_wrong_type;
goto out_put_dst;
}
NFSD: Implement the COPY call I only implemented the sync version of this call, since it's the easiest. I can simply call vfs_copy_range() and have the vfs do the right thing for the filesystem being exported. Signed-off-by: Anna Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-09-08 03:57:30 +08:00
out:
return status;
out_put_dst:
nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp. Since we now presume that the struct file will be persistent in most cases, we can stop fiddling with the raparms in the read code. This also means that we don't really care about the rd_tmp_file field anymore. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-08-19 02:18:54 +08:00
nfsd_file_put(*dst);
nfsd: zero out pointers after putting nfsd_files on COPY setup error At first, I thought this might be a source of nfsd_file overputs, but the current callers seem to avoid an extra put when nfsd4_verify_copy returns an error. Still, it's "bad form" to leave the pointers filled out when we don't have a reference to them anymore, and that might lead to bugs later. Zero them out as a defensive coding measure. Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-01-18 03:38:30 +08:00
*dst = NULL;
NFSD: Implement the COPY call I only implemented the sync version of this call, since it's the easiest. I can simply call vfs_copy_range() and have the vfs do the right thing for the filesystem being exported. Signed-off-by: Anna Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-09-08 03:57:30 +08:00
out_put_src:
nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp. Since we now presume that the struct file will be persistent in most cases, we can stop fiddling with the raparms in the read code. This also means that we don't really care about the rd_tmp_file field anymore. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-08-19 02:18:54 +08:00
nfsd_file_put(*src);
nfsd: zero out pointers after putting nfsd_files on COPY setup error At first, I thought this might be a source of nfsd_file overputs, but the current callers seem to avoid an extra put when nfsd4_verify_copy returns an error. Still, it's "bad form" to leave the pointers filled out when we don't have a reference to them anymore, and that might lead to bugs later. Zero them out as a defensive coding measure. Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-01-18 03:38:30 +08:00
*src = NULL;
NFSD: Implement the COPY call I only implemented the sync version of this call, since it's the easiest. I can simply call vfs_copy_range() and have the vfs do the right thing for the filesystem being exported. Signed-off-by: Anna Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-09-08 03:57:30 +08:00
goto out;
}
static __be32
nfsd4_clone(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
NFSD: Implement the COPY call I only implemented the sync version of this call, since it's the easiest. I can simply call vfs_copy_range() and have the vfs do the right thing for the filesystem being exported. Signed-off-by: Anna Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-09-08 03:57:30 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_clone *clone = &u->clone;
nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp. Since we now presume that the struct file will be persistent in most cases, we can stop fiddling with the raparms in the read code. This also means that we don't really care about the rd_tmp_file field anymore. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-08-19 02:18:54 +08:00
struct nfsd_file *src, *dst;
NFSD: Implement the COPY call I only implemented the sync version of this call, since it's the easiest. I can simply call vfs_copy_range() and have the vfs do the right thing for the filesystem being exported. Signed-off-by: Anna Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-09-08 03:57:30 +08:00
__be32 status;
status = nfsd4_verify_copy(rqstp, cstate, &clone->cl_src_stateid, &src,
&clone->cl_dst_stateid, &dst);
if (status)
goto out;
nfsd: Add a tracepoint for errors in nfsd4_clone_file_range() Since a clone error commit can cause the boot verifier to change, we should trace those errors. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: Addressed a checkpatch.pl splat in fs/nfsd/vfs.h ]
2021-12-19 09:38:00 +08:00
status = nfsd4_clone_file_range(rqstp, src, clone->cl_src_pos,
nfsd: Pass the nfsd_file as arguments to nfsd4_clone_file_range() Needed in order to fix exclusion w.r.t. writes. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2020-01-07 02:40:32 +08:00
dst, clone->cl_dst_pos, clone->cl_count,
nfsd: Ensure CLONE persists data and metadata changes to the target file The NFSv4.2 CLONE operation has implicit persistence requirements on the target file, since there is no protocol requirement that the client issue a separate operation to persist data. For that reason, we should call vfs_fsync_range() on the destination file after a successful call to vfs_clone_file_range(). Fixes: ffa0160a1039 ("nfsd: implement the NFSv4.2 CLONE operation") Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Cc: stable@vger.kernel.org # v4.5+ Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-11-28 06:05:51 +08:00
EX_ISSYNC(cstate->current_fh.fh_export));
nfsd: implement the NFSv4.2 CLONE operation This is basically a remote version of the btrfs CLONE operation, so the implementation is fairly trivial. Made even more trivial by stealing the XDR code and general framework Anna Schumaker's COPY prototype. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 19:59:52 +08:00
nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp. Since we now presume that the struct file will be persistent in most cases, we can stop fiddling with the raparms in the read code. This also means that we don't really care about the rd_tmp_file field anymore. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-08-19 02:18:54 +08:00
nfsd_file_put(dst);
nfsd_file_put(src);
nfsd: implement the NFSv4.2 CLONE operation This is basically a remote version of the btrfs CLONE operation, so the implementation is fairly trivial. Made even more trivial by stealing the XDR code and general framework Anna Schumaker's COPY prototype. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 19:59:52 +08:00
out:
return status;
}
NFSD: Make nfs4_put_copy() static Clean up: All call sites are in fs/nfsd/nfs4proc.c. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:35 +08:00
static void nfs4_put_copy(struct nfsd4_copy *copy)
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
{
if (!refcount_dec_and_test(&copy->refcount))
return;
NFSD: Limit the number of concurrent async COPY operations Nothing appears to limit the number of concurrent async COPY operations that clients can start. In addition, AFAICT each async COPY can copy an unlimited number of 4MB chunks, so can run for a long time. Thus IMO async COPY can become a DoS vector. Add a restriction mechanism that bounds the number of concurrent background COPY operations. Start simple and try to be fair -- this patch implements a per-namespace limit. An async COPY request that occurs while this limit is exceeded gets NFS4ERR_DELAY. The requesting client can choose to send the request again after a delay or fall back to a traditional read/write style copy. If there is need to make the mechanism more sophisticated, we can visit that in future patches. Cc: stable@vger.kernel.org Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-08-29 01:40:04 +08:00
atomic_dec(&copy->cp_nn->pending_async_copies);
NFSD: Shrink size of struct nfsd4_copy struct nfsd4_copy is part of struct nfsd4_op, which resides in an 8-element array. sizeof(struct nfsd4_op): Before: /* size: 1696, cachelines: 27, members: 5 */ After: /* size: 672, cachelines: 11, members: 5 */ Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:22 +08:00
kfree(copy->cp_src);
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
kfree(copy);
}
static void nfsd4_stop_copy(struct nfsd4_copy *copy)
{
NFSD: Replace boolean fields in struct nfsd4_copy Clean up: saves 8 bytes, and we can replace check_and_set_stop_copy() with an atomic bitop. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:41 +08:00
if (!test_and_set_bit(NFSD4_COPY_F_STOPPED, &copy->cp_flags))
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
kthread_stop(copy->copy_task);
nfs4_put_copy(copy);
}
static struct nfsd4_copy *nfsd4_get_copy(struct nfs4_client *clp)
{
struct nfsd4_copy *copy = NULL;
spin_lock(&clp->async_lock);
if (!list_empty(&clp->async_copies)) {
copy = list_first_entry(&clp->async_copies, struct nfsd4_copy,
copies);
refcount_inc(&copy->refcount);
}
spin_unlock(&clp->async_lock);
return copy;
}
void nfsd4_shutdown_copy(struct nfs4_client *clp)
{
struct nfsd4_copy *copy;
while ((copy = nfsd4_get_copy(clp)) != NULL)
nfsd4_stop_copy(copy);
}
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
#ifdef CONFIG_NFSD_V4_2_INTER_SSC
extern struct file *nfs42_ssc_open(struct vfsmount *ss_mnt,
struct nfs_fh *src_fh,
nfs4_stateid *stateid);
extern void nfs42_ssc_close(struct file *filep);
extern void nfs_sb_deactive(struct super_block *sb);
#define NFSD42_INTERSSC_MOUNTOPS "vers=4.2,addr=%s,sec=sys"
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
/*
* setup a work entry in the ssc delayed unmount list.
*/
nfsd: fix kernel test robot warning in SSC code Fix by initializing pointer nfsd4_ssc_umount_item with NULL instead of 0. Replace return value of nfsd4_ssc_setup_dul with __be32 instead of int. Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-06-04 08:02:26 +08:00
static __be32 nfsd4_ssc_setup_dul(struct nfsd_net *nn, char *ipaddr,
SUNRPC: change how svc threads are asked to exit. svc threads are currently stopped using kthread_stop(). This requires identifying a specific thread. However we don't care which thread stops, just as long as one does. So instead, set a flag in the svc_pool to say that a thread needs to die, and have each thread check this flag instead of calling kthread_should_stop(). The first thread to find and clear this flag then moves towards exiting. This removes an explicit dependency on sp_all_threads which will make a future patch simpler. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-09-11 22:39:04 +08:00
struct nfsd4_ssc_umount_item **nsui,
struct svc_rqst *rqstp)
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
{
NFSD: Initialize pointer ni with NULL and not plain integer 0 Pointer ni is being initialized with plain integer zero. Fix this by initializing with NULL. Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-09-26 06:58:41 +08:00
struct nfsd4_ssc_umount_item *ni = NULL;
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
struct nfsd4_ssc_umount_item *work = NULL;
struct nfsd4_ssc_umount_item *tmp;
DEFINE_WAIT(wait);
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
__be32 status = 0;
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
*nsui = NULL;
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
work = kzalloc(sizeof(*work), GFP_KERNEL);
try_again:
spin_lock(&nn->nfsd_ssc_lock);
list_for_each_entry_safe(ni, tmp, &nn->nfsd_ssc_mount_list, nsui_list) {
if (strncmp(ni->nsui_ipaddr, ipaddr, sizeof(ni->nsui_ipaddr)))
continue;
/* found a match */
if (ni->nsui_busy) {
/* wait - and try again */
nfsd: don't allow nfsd threads to be signalled. The original implementation of nfsd used signals to stop threads during shutdown. In Linux 2.3.46pre5 nfsd gained the ability to shutdown threads internally it if was asked to run "0" threads. After this user-space transitioned to using "rpc.nfsd 0" to stop nfsd and sending signals to threads was no longer an important part of the API. In commit 3ebdbe5203a8 ("SUNRPC: discard svo_setup and rename svc_set_num_threads_sync()") (v5.17-rc1~75^2~41) we finally removed the use of signals for stopping threads, using kthread_stop() instead. This patch makes the "obvious" next step and removes the ability to signal nfsd threads - or any svc threads. nfsd stops allowing signals and we don't check for their delivery any more. This will allow for some simplification in later patches. A change worth noting is in nfsd4_ssc_setup_dul(). There was previously a signal_pending() check which would only succeed when the thread was being shut down. It should really have tested kthread_should_stop() as well. Now it just does the latter, not the former. Signed-off-by: NeilBrown <neilb@suse.de> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-07-18 14:38:08 +08:00
prepare_to_wait(&nn->nfsd_ssc_waitq, &wait, TASK_IDLE);
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
spin_unlock(&nn->nfsd_ssc_lock);
/* allow 20secs for mount/unmount for now - revisit */
SUNRPC: change how svc threads are asked to exit. svc threads are currently stopped using kthread_stop(). This requires identifying a specific thread. However we don't care which thread stops, just as long as one does. So instead, set a flag in the svc_pool to say that a thread needs to die, and have each thread check this flag instead of calling kthread_should_stop(). The first thread to find and clear this flag then moves towards exiting. This removes an explicit dependency on sp_all_threads which will make a future patch simpler. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-09-11 22:39:04 +08:00
if (svc_thread_should_stop(rqstp) ||
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
(schedule_timeout(20*HZ) == 0)) {
NFSD: fix use-after-free in nfsd4_ssc_setup_dul() If signal_pending() returns true, schedule_timeout() will not be executed, causing the waiting task to remain in the wait queue. Fixed by adding a call to finish_wait(), which ensures that the waiting task will always be removed from the wait queue. Fixes: f4e44b393389 ("NFSD: delay unmount source's export after inter-server copy completed.") Signed-off-by: Xingyuan Mo <hdthky0@gmail.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-01-12 00:24:53 +08:00
finish_wait(&nn->nfsd_ssc_waitq, &wait);
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
kfree(work);
return nfserr_eagain;
}
finish_wait(&nn->nfsd_ssc_waitq, &wait);
goto try_again;
}
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
*nsui = ni;
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
refcount_inc(&ni->nsui_refcnt);
spin_unlock(&nn->nfsd_ssc_lock);
kfree(work);
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
/* return vfsmount in (*nsui)->nsui_vfsmount */
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
return 0;
}
if (work) {
NFSD: move from strlcpy with unused retval to strscpy Follow the advice of the below link and prefer 'strscpy' in this subsystem. Conversion is 1:1 because the return value is not used. Generated by a coccinelle script. Link: https://lore.kernel.org/r/CAHk-=wgfRnXz0W3D37d01q3JFkr_i_uTL=V6A6G1oUZcprmknw@mail.gmail.com/ Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-08-19 05:01:14 +08:00
strscpy(work->nsui_ipaddr, ipaddr, sizeof(work->nsui_ipaddr) - 1);
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
refcount_set(&work->nsui_refcnt, 2);
work->nsui_busy = true;
list_add_tail(&work->nsui_list, &nn->nfsd_ssc_mount_list);
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
*nsui = work;
} else
status = nfserr_resource;
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
spin_unlock(&nn->nfsd_ssc_lock);
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
return status;
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
}
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
static void nfsd4_ssc_update_dul(struct nfsd_net *nn,
struct nfsd4_ssc_umount_item *nsui,
struct vfsmount *ss_mnt)
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
{
spin_lock(&nn->nfsd_ssc_lock);
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
nsui->nsui_vfsmount = ss_mnt;
nsui->nsui_busy = false;
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
wake_up_all(&nn->nfsd_ssc_waitq);
spin_unlock(&nn->nfsd_ssc_lock);
}
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
static void nfsd4_ssc_cancel_dul(struct nfsd_net *nn,
struct nfsd4_ssc_umount_item *nsui)
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
{
spin_lock(&nn->nfsd_ssc_lock);
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
list_del(&nsui->nsui_list);
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
wake_up_all(&nn->nfsd_ssc_waitq);
spin_unlock(&nn->nfsd_ssc_lock);
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
kfree(nsui);
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
}
NFSD: Fix improperly-formatted Doxygen comments fs/nfsd/nfsctl.c:256: warning: Function parameter or member 'file' not described in 'write_unlock_ip' fs/nfsd/nfsctl.c:256: warning: Function parameter or member 'buf' not described in 'write_unlock_ip' fs/nfsd/nfsctl.c:256: warning: Function parameter or member 'size' not described in 'write_unlock_ip' fs/nfsd/nfsctl.c:295: warning: Function parameter or member 'file' not described in 'write_unlock_fs' fs/nfsd/nfsctl.c:295: warning: Function parameter or member 'buf' not described in 'write_unlock_fs' fs/nfsd/nfsctl.c:295: warning: Function parameter or member 'size' not described in 'write_unlock_fs' fs/nfsd/nfsctl.c:352: warning: Function parameter or member 'file' not described in 'write_filehandle' fs/nfsd/nfsctl.c:352: warning: Function parameter or member 'buf' not described in 'write_filehandle' fs/nfsd/nfsctl.c:352: warning: Function parameter or member 'size' not described in 'write_filehandle' fs/nfsd/nfsctl.c:434: warning: Function parameter or member 'file' not described in 'write_threads' fs/nfsd/nfsctl.c:434: warning: Function parameter or member 'buf' not described in 'write_threads' fs/nfsd/nfsctl.c:434: warning: Function parameter or member 'size' not described in 'write_threads' fs/nfsd/nfsctl.c:478: warning: Function parameter or member 'file' not described in 'write_pool_threads' fs/nfsd/nfsctl.c:478: warning: Function parameter or member 'buf' not described in 'write_pool_threads' fs/nfsd/nfsctl.c:478: warning: Function parameter or member 'size' not described in 'write_pool_threads' fs/nfsd/nfsctl.c:697: warning: Function parameter or member 'file' not described in 'write_versions' fs/nfsd/nfsctl.c:697: warning: Function parameter or member 'buf' not described in 'write_versions' fs/nfsd/nfsctl.c:697: warning: Function parameter or member 'size' not described in 'write_versions' fs/nfsd/nfsctl.c:858: warning: Function parameter or member 'file' not described in 'write_ports' fs/nfsd/nfsctl.c:858: warning: Function parameter or member 'buf' not described in 'write_ports' fs/nfsd/nfsctl.c:858: warning: Function parameter or member 'size' not described in 'write_ports' fs/nfsd/nfsctl.c:892: warning: Function parameter or member 'file' not described in 'write_maxblksize' fs/nfsd/nfsctl.c:892: warning: Function parameter or member 'buf' not described in 'write_maxblksize' fs/nfsd/nfsctl.c:892: warning: Function parameter or member 'size' not described in 'write_maxblksize' fs/nfsd/nfsctl.c:941: warning: Function parameter or member 'file' not described in 'write_maxconn' fs/nfsd/nfsctl.c:941: warning: Function parameter or member 'buf' not described in 'write_maxconn' fs/nfsd/nfsctl.c:941: warning: Function parameter or member 'size' not described in 'write_maxconn' fs/nfsd/nfsctl.c:1023: warning: Function parameter or member 'file' not described in 'write_leasetime' fs/nfsd/nfsctl.c:1023: warning: Function parameter or member 'buf' not described in 'write_leasetime' fs/nfsd/nfsctl.c:1023: warning: Function parameter or member 'size' not described in 'write_leasetime' fs/nfsd/nfsctl.c:1039: warning: Function parameter or member 'file' not described in 'write_gracetime' fs/nfsd/nfsctl.c:1039: warning: Function parameter or member 'buf' not described in 'write_gracetime' fs/nfsd/nfsctl.c:1039: warning: Function parameter or member 'size' not described in 'write_gracetime' fs/nfsd/nfsctl.c:1094: warning: Function parameter or member 'file' not described in 'write_recoverydir' fs/nfsd/nfsctl.c:1094: warning: Function parameter or member 'buf' not described in 'write_recoverydir' fs/nfsd/nfsctl.c:1094: warning: Function parameter or member 'size' not described in 'write_recoverydir' fs/nfsd/nfsctl.c:1125: warning: Function parameter or member 'file' not described in 'write_v4_end_grace' fs/nfsd/nfsctl.c:1125: warning: Function parameter or member 'buf' not described in 'write_v4_end_grace' fs/nfsd/nfsctl.c:1125: warning: Function parameter or member 'size' not described in 'write_v4_end_grace' fs/nfsd/nfs4proc.c:1164: warning: Function parameter or member 'nss' not described in 'nfsd4_interssc_connect' fs/nfsd/nfs4proc.c:1164: warning: Function parameter or member 'rqstp' not described in 'nfsd4_interssc_connect' fs/nfsd/nfs4proc.c:1164: warning: Function parameter or member 'mount' not described in 'nfsd4_interssc_connect' fs/nfsd/nfs4proc.c:1262: warning: Function parameter or member 'rqstp' not described in 'nfsd4_setup_inter_ssc' fs/nfsd/nfs4proc.c:1262: warning: Function parameter or member 'cstate' not described in 'nfsd4_setup_inter_ssc' fs/nfsd/nfs4proc.c:1262: warning: Function parameter or member 'copy' not described in 'nfsd4_setup_inter_ssc' fs/nfsd/nfs4proc.c:1262: warning: Function parameter or member 'mount' not described in 'nfsd4_setup_inter_ssc' Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-04-07 04:01:42 +08:00
/*
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
* Support one copy source server for now.
*/
static __be32
nfsd4_interssc_connect(struct nl4_server *nss, struct svc_rqst *rqstp,
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
struct nfsd4_ssc_umount_item **nsui)
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
{
struct file_system_type *type;
struct vfsmount *ss_mnt;
struct nfs42_netaddr *naddr;
struct sockaddr_storage tmp_addr;
size_t tmp_addrlen, match_netid_len = 3;
char *startsep = "", *endsep = "", *match_netid = "tcp";
char *ipaddr, *dev_name, *raw_data;
NFSD fix nfserro errno mismatch There is mismatch between __be32 and u32 in nfserr and errno. Reported-by: kbuild test robot <lkp@intel.com> Fixes: d5e54eeb0e3d ("NFSD add nfs4 inter ssc to nfsd4_copy") Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-12-05 04:13:53 +08:00
int len, raw_len;
__be32 status = nfserr_inval;
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
naddr = &nss->u.nl4_addr;
tmp_addrlen = rpc_uaddr2sockaddr(SVC_NET(rqstp), naddr->addr,
naddr->addr_len,
(struct sockaddr *)&tmp_addr,
sizeof(tmp_addr));
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
*nsui = NULL;
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
if (tmp_addrlen == 0)
goto out_err;
if (tmp_addr.ss_family == AF_INET6) {
startsep = "[";
endsep = "]";
match_netid = "tcp6";
match_netid_len = 4;
}
if (naddr->netid_len != match_netid_len ||
strncmp(naddr->netid, match_netid, naddr->netid_len))
goto out_err;
/* Construct the raw data for the vfs_kern_mount call */
len = RPC_MAX_ADDRBUFLEN + 1;
ipaddr = kzalloc(len, GFP_KERNEL);
if (!ipaddr)
goto out_err;
rpc_ntop((struct sockaddr *)&tmp_addr, ipaddr, len);
/* 2 for ipv6 endsep and startsep. 3 for ":/" and trailing '/0'*/
raw_len = strlen(NFSD42_INTERSSC_MOUNTOPS) + strlen(ipaddr);
raw_data = kzalloc(raw_len, GFP_KERNEL);
if (!raw_data)
goto out_free_ipaddr;
snprintf(raw_data, raw_len, NFSD42_INTERSSC_MOUNTOPS, ipaddr);
NFSD fix nfserro errno mismatch There is mismatch between __be32 and u32 in nfserr and errno. Reported-by: kbuild test robot <lkp@intel.com> Fixes: d5e54eeb0e3d ("NFSD add nfs4 inter ssc to nfsd4_copy") Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-12-05 04:13:53 +08:00
status = nfserr_nodev;
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
type = get_fs_type("nfs");
if (!type)
goto out_free_rawdata;
/* Set the server:<export> for the vfs_kern_mount call */
dev_name = kzalloc(len + 5, GFP_KERNEL);
if (!dev_name)
goto out_free_rawdata;
snprintf(dev_name, len + 5, "%s%s%s:/", startsep, ipaddr, endsep);
SUNRPC: change how svc threads are asked to exit. svc threads are currently stopped using kthread_stop(). This requires identifying a specific thread. However we don't care which thread stops, just as long as one does. So instead, set a flag in the svc_pool to say that a thread needs to die, and have each thread check this flag instead of calling kthread_should_stop(). The first thread to find and clear this flag then moves towards exiting. This removes an explicit dependency on sp_all_threads which will make a future patch simpler. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-09-11 22:39:04 +08:00
status = nfsd4_ssc_setup_dul(nn, ipaddr, nsui, rqstp);
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
if (status)
goto out_free_devname;
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
if ((*nsui)->nsui_vfsmount)
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
goto out_done;
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
/* Use an 'internal' mount: SB_KERNMOUNT -> MNT_INTERNAL */
ss_mnt = vfs_kern_mount(type, SB_KERNMOUNT, dev_name, raw_data);
module_put(type->owner);
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
if (IS_ERR(ss_mnt)) {
NFSD: Fix error return code in nfsd4_interssc_connect() 'status' has been overwritten to 0 after nfsd4_ssc_setup_dul(), this cause 0 will be return in vfs_kern_mount() error case. Fix to return nfserr_nodev in this error. Fixes: f4e44b393389 ("NFSD: delay unmount source's export after inter-server copy completed.") Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-06-04 18:12:37 +08:00
status = nfserr_nodev;
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
nfsd4_ssc_cancel_dul(nn, *nsui);
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
goto out_free_devname;
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
}
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
nfsd4_ssc_update_dul(nn, *nsui, ss_mnt);
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
out_done:
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
status = 0;
out_free_devname:
kfree(dev_name);
out_free_rawdata:
kfree(raw_data);
out_free_ipaddr:
kfree(ipaddr);
out_err:
return status;
}
NFSD: Fix improperly-formatted Doxygen comments fs/nfsd/nfsctl.c:256: warning: Function parameter or member 'file' not described in 'write_unlock_ip' fs/nfsd/nfsctl.c:256: warning: Function parameter or member 'buf' not described in 'write_unlock_ip' fs/nfsd/nfsctl.c:256: warning: Function parameter or member 'size' not described in 'write_unlock_ip' fs/nfsd/nfsctl.c:295: warning: Function parameter or member 'file' not described in 'write_unlock_fs' fs/nfsd/nfsctl.c:295: warning: Function parameter or member 'buf' not described in 'write_unlock_fs' fs/nfsd/nfsctl.c:295: warning: Function parameter or member 'size' not described in 'write_unlock_fs' fs/nfsd/nfsctl.c:352: warning: Function parameter or member 'file' not described in 'write_filehandle' fs/nfsd/nfsctl.c:352: warning: Function parameter or member 'buf' not described in 'write_filehandle' fs/nfsd/nfsctl.c:352: warning: Function parameter or member 'size' not described in 'write_filehandle' fs/nfsd/nfsctl.c:434: warning: Function parameter or member 'file' not described in 'write_threads' fs/nfsd/nfsctl.c:434: warning: Function parameter or member 'buf' not described in 'write_threads' fs/nfsd/nfsctl.c:434: warning: Function parameter or member 'size' not described in 'write_threads' fs/nfsd/nfsctl.c:478: warning: Function parameter or member 'file' not described in 'write_pool_threads' fs/nfsd/nfsctl.c:478: warning: Function parameter or member 'buf' not described in 'write_pool_threads' fs/nfsd/nfsctl.c:478: warning: Function parameter or member 'size' not described in 'write_pool_threads' fs/nfsd/nfsctl.c:697: warning: Function parameter or member 'file' not described in 'write_versions' fs/nfsd/nfsctl.c:697: warning: Function parameter or member 'buf' not described in 'write_versions' fs/nfsd/nfsctl.c:697: warning: Function parameter or member 'size' not described in 'write_versions' fs/nfsd/nfsctl.c:858: warning: Function parameter or member 'file' not described in 'write_ports' fs/nfsd/nfsctl.c:858: warning: Function parameter or member 'buf' not described in 'write_ports' fs/nfsd/nfsctl.c:858: warning: Function parameter or member 'size' not described in 'write_ports' fs/nfsd/nfsctl.c:892: warning: Function parameter or member 'file' not described in 'write_maxblksize' fs/nfsd/nfsctl.c:892: warning: Function parameter or member 'buf' not described in 'write_maxblksize' fs/nfsd/nfsctl.c:892: warning: Function parameter or member 'size' not described in 'write_maxblksize' fs/nfsd/nfsctl.c:941: warning: Function parameter or member 'file' not described in 'write_maxconn' fs/nfsd/nfsctl.c:941: warning: Function parameter or member 'buf' not described in 'write_maxconn' fs/nfsd/nfsctl.c:941: warning: Function parameter or member 'size' not described in 'write_maxconn' fs/nfsd/nfsctl.c:1023: warning: Function parameter or member 'file' not described in 'write_leasetime' fs/nfsd/nfsctl.c:1023: warning: Function parameter or member 'buf' not described in 'write_leasetime' fs/nfsd/nfsctl.c:1023: warning: Function parameter or member 'size' not described in 'write_leasetime' fs/nfsd/nfsctl.c:1039: warning: Function parameter or member 'file' not described in 'write_gracetime' fs/nfsd/nfsctl.c:1039: warning: Function parameter or member 'buf' not described in 'write_gracetime' fs/nfsd/nfsctl.c:1039: warning: Function parameter or member 'size' not described in 'write_gracetime' fs/nfsd/nfsctl.c:1094: warning: Function parameter or member 'file' not described in 'write_recoverydir' fs/nfsd/nfsctl.c:1094: warning: Function parameter or member 'buf' not described in 'write_recoverydir' fs/nfsd/nfsctl.c:1094: warning: Function parameter or member 'size' not described in 'write_recoverydir' fs/nfsd/nfsctl.c:1125: warning: Function parameter or member 'file' not described in 'write_v4_end_grace' fs/nfsd/nfsctl.c:1125: warning: Function parameter or member 'buf' not described in 'write_v4_end_grace' fs/nfsd/nfsctl.c:1125: warning: Function parameter or member 'size' not described in 'write_v4_end_grace' fs/nfsd/nfs4proc.c:1164: warning: Function parameter or member 'nss' not described in 'nfsd4_interssc_connect' fs/nfsd/nfs4proc.c:1164: warning: Function parameter or member 'rqstp' not described in 'nfsd4_interssc_connect' fs/nfsd/nfs4proc.c:1164: warning: Function parameter or member 'mount' not described in 'nfsd4_interssc_connect' fs/nfsd/nfs4proc.c:1262: warning: Function parameter or member 'rqstp' not described in 'nfsd4_setup_inter_ssc' fs/nfsd/nfs4proc.c:1262: warning: Function parameter or member 'cstate' not described in 'nfsd4_setup_inter_ssc' fs/nfsd/nfs4proc.c:1262: warning: Function parameter or member 'copy' not described in 'nfsd4_setup_inter_ssc' fs/nfsd/nfs4proc.c:1262: warning: Function parameter or member 'mount' not described in 'nfsd4_setup_inter_ssc' Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-04-07 04:01:42 +08:00
/*
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
* Verify COPY destination stateid.
NFSD: Fix improperly-formatted Doxygen comments fs/nfsd/nfsctl.c:256: warning: Function parameter or member 'file' not described in 'write_unlock_ip' fs/nfsd/nfsctl.c:256: warning: Function parameter or member 'buf' not described in 'write_unlock_ip' fs/nfsd/nfsctl.c:256: warning: Function parameter or member 'size' not described in 'write_unlock_ip' fs/nfsd/nfsctl.c:295: warning: Function parameter or member 'file' not described in 'write_unlock_fs' fs/nfsd/nfsctl.c:295: warning: Function parameter or member 'buf' not described in 'write_unlock_fs' fs/nfsd/nfsctl.c:295: warning: Function parameter or member 'size' not described in 'write_unlock_fs' fs/nfsd/nfsctl.c:352: warning: Function parameter or member 'file' not described in 'write_filehandle' fs/nfsd/nfsctl.c:352: warning: Function parameter or member 'buf' not described in 'write_filehandle' fs/nfsd/nfsctl.c:352: warning: Function parameter or member 'size' not described in 'write_filehandle' fs/nfsd/nfsctl.c:434: warning: Function parameter or member 'file' not described in 'write_threads' fs/nfsd/nfsctl.c:434: warning: Function parameter or member 'buf' not described in 'write_threads' fs/nfsd/nfsctl.c:434: warning: Function parameter or member 'size' not described in 'write_threads' fs/nfsd/nfsctl.c:478: warning: Function parameter or member 'file' not described in 'write_pool_threads' fs/nfsd/nfsctl.c:478: warning: Function parameter or member 'buf' not described in 'write_pool_threads' fs/nfsd/nfsctl.c:478: warning: Function parameter or member 'size' not described in 'write_pool_threads' fs/nfsd/nfsctl.c:697: warning: Function parameter or member 'file' not described in 'write_versions' fs/nfsd/nfsctl.c:697: warning: Function parameter or member 'buf' not described in 'write_versions' fs/nfsd/nfsctl.c:697: warning: Function parameter or member 'size' not described in 'write_versions' fs/nfsd/nfsctl.c:858: warning: Function parameter or member 'file' not described in 'write_ports' fs/nfsd/nfsctl.c:858: warning: Function parameter or member 'buf' not described in 'write_ports' fs/nfsd/nfsctl.c:858: warning: Function parameter or member 'size' not described in 'write_ports' fs/nfsd/nfsctl.c:892: warning: Function parameter or member 'file' not described in 'write_maxblksize' fs/nfsd/nfsctl.c:892: warning: Function parameter or member 'buf' not described in 'write_maxblksize' fs/nfsd/nfsctl.c:892: warning: Function parameter or member 'size' not described in 'write_maxblksize' fs/nfsd/nfsctl.c:941: warning: Function parameter or member 'file' not described in 'write_maxconn' fs/nfsd/nfsctl.c:941: warning: Function parameter or member 'buf' not described in 'write_maxconn' fs/nfsd/nfsctl.c:941: warning: Function parameter or member 'size' not described in 'write_maxconn' fs/nfsd/nfsctl.c:1023: warning: Function parameter or member 'file' not described in 'write_leasetime' fs/nfsd/nfsctl.c:1023: warning: Function parameter or member 'buf' not described in 'write_leasetime' fs/nfsd/nfsctl.c:1023: warning: Function parameter or member 'size' not described in 'write_leasetime' fs/nfsd/nfsctl.c:1039: warning: Function parameter or member 'file' not described in 'write_gracetime' fs/nfsd/nfsctl.c:1039: warning: Function parameter or member 'buf' not described in 'write_gracetime' fs/nfsd/nfsctl.c:1039: warning: Function parameter or member 'size' not described in 'write_gracetime' fs/nfsd/nfsctl.c:1094: warning: Function parameter or member 'file' not described in 'write_recoverydir' fs/nfsd/nfsctl.c:1094: warning: Function parameter or member 'buf' not described in 'write_recoverydir' fs/nfsd/nfsctl.c:1094: warning: Function parameter or member 'size' not described in 'write_recoverydir' fs/nfsd/nfsctl.c:1125: warning: Function parameter or member 'file' not described in 'write_v4_end_grace' fs/nfsd/nfsctl.c:1125: warning: Function parameter or member 'buf' not described in 'write_v4_end_grace' fs/nfsd/nfsctl.c:1125: warning: Function parameter or member 'size' not described in 'write_v4_end_grace' fs/nfsd/nfs4proc.c:1164: warning: Function parameter or member 'nss' not described in 'nfsd4_interssc_connect' fs/nfsd/nfs4proc.c:1164: warning: Function parameter or member 'rqstp' not described in 'nfsd4_interssc_connect' fs/nfsd/nfs4proc.c:1164: warning: Function parameter or member 'mount' not described in 'nfsd4_interssc_connect' fs/nfsd/nfs4proc.c:1262: warning: Function parameter or member 'rqstp' not described in 'nfsd4_setup_inter_ssc' fs/nfsd/nfs4proc.c:1262: warning: Function parameter or member 'cstate' not described in 'nfsd4_setup_inter_ssc' fs/nfsd/nfs4proc.c:1262: warning: Function parameter or member 'copy' not described in 'nfsd4_setup_inter_ssc' fs/nfsd/nfs4proc.c:1262: warning: Function parameter or member 'mount' not described in 'nfsd4_setup_inter_ssc' Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-04-07 04:01:42 +08:00
*
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
* Connect to the source server with NFSv4.1.
* Create the source struct file for nfsd_copy_range.
* Called with COPY cstate:
* SAVED_FH: source filehandle
* CURRENT_FH: destination filehandle
*/
static __be32
nfsd4_setup_inter_ssc(struct svc_rqst *rqstp,
struct nfsd4_compound_state *cstate,
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
struct nfsd4_copy *copy)
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
{
struct svc_fh *s_fh = NULL;
stateid_t *s_stid = &copy->cp_src_stateid;
NFSD fix nfserro errno mismatch There is mismatch between __be32 and u32 in nfserr and errno. Reported-by: kbuild test robot <lkp@intel.com> Fixes: d5e54eeb0e3d ("NFSD add nfs4 inter ssc to nfsd4_copy") Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-12-05 04:13:53 +08:00
__be32 status = nfserr_inval;
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
/* Verify the destination stateid and set dst struct file*/
status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
&copy->cp_dst_stateid,
WR_STATE, &copy->nf_dst, NULL);
if (status)
goto out;
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
status = nfsd4_interssc_connect(copy->cp_src, rqstp, &copy->ss_nsui);
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
if (status)
goto out;
s_fh = &cstate->save_fh;
copy->c_fh.size = s_fh->fh_handle.fh_size;
NFSD: simplify struct nfsfh Most of the fields in 'struct knfsd_fh' are 2 levels deep (a union and a struct) and are accessed using macros like: #define fh_FOO fh_base.fh_new.fb_FOO This patch makes the union and struct anonymous, so that "fh_FOO" can be a name directly within 'struct knfsd_fh' and the #defines aren't needed. The file handle as a whole is sometimes accessed as "fh_base" or "fh_base.fh_pad", neither of which are particularly helpful names. As the struct holding the filehandle is now anonymous, we cannot use the name of that, so we union it with 'fh_raw' and use that where the raw filehandle is needed. fh_raw also ensure the structure is large enough for the largest possible filehandle. fh_raw is a 'char' array, removing any need to cast it for memcpy etc. SVCFH_fmt() is simplified using the "%ph" printk format. This changes the appearance of filehandles in dprintk() debugging, making them a little more precise. Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-09-02 09:16:32 +08:00
memcpy(copy->c_fh.data, &s_fh->fh_handle.fh_raw, copy->c_fh.size);
NFSD: fix seqid in copy stateid s_stid->si_generation is a u32, copy->stateid.seqid is a __be32, so we should be byte-swapping here if necessary. This effectively undoes the byte-swap performed when reading s_stid->s_generation in nfsd4_decode_copy(). Without this second swap, the stateid we sent to the source in READ could be different from the one the client provided us in the COPY. We didn't spot this in testing since our implementation always uses a 0 in the seqid field. But other implementations might not do that. You'd think we should just skip the byte-swapping entirely, but the s_stid field can be used for either our own stateids (in the intra-server case) or foreign stateids (in the inter-server case), and the former are interpreted by us and need byte-swapping. Reported-by: kbuild test robot <lkp@intel.com> Fixes: d5e54eeb0e3d ("NFSD add nfs4 inter ssc to nfsd4_copy") Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-12-05 04:13:53 +08:00
copy->stateid.seqid = cpu_to_be32(s_stid->si_generation);
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
memcpy(copy->stateid.other, (void *)&s_stid->si_opaque,
sizeof(stateid_opaque_t));
status = 0;
out:
return status;
}
static void
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
nfsd4_cleanup_inter_ssc(struct nfsd4_ssc_umount_item *nsui, struct file *filp,
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
struct nfsd_file *dst)
{
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
struct nfsd_net *nn = net_generic(dst->nf_net, nfsd_net_id);
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
long timeout = msecs_to_jiffies(nfsd4_ssc_umount_timeout);
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
NFSD: Refactor nfsd4_cleanup_inter_ssc() (1/2) The @src parameter is sometimes a pointer to a struct nfsd_file and sometimes a pointer to struct file hiding in a phony struct nfsd_file. Refactor nfsd4_cleanup_inter_ssc() so the @src parameter is always an explicit struct file. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:47 +08:00
nfs42_ssc_close(filp);
fput(filp);
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
spin_lock(&nn->nfsd_ssc_lock);
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
list_del(&nsui->nsui_list);
/*
* vfsmount can be shared by multiple exports,
* decrement refcnt. If the count drops to 1 it
* will be unmounted when nsui_expire expires.
*/
refcount_dec(&nsui->nsui_refcnt);
nsui->nsui_expire = jiffies + timeout;
list_add_tail(&nsui->nsui_list, &nn->nfsd_ssc_mount_list);
NFSD: delay unmount source's export after inter-server copy completed. Currently the source's export is mounted and unmounted on every inter-server copy operation. This patch is an enhancement to delay the unmount of the source export for a certain period of time to eliminate the mount and unmount overhead on subsequent copy operations. After a copy operation completes, a work entry is added to the delayed unmount list with an expiration time. This list is serviced by the laundromat thread to unmount the export of the expired entries. Each time the export is being used again, its expiration time is extended and the entry is re-inserted to the tail of the list. The unmount task and the mount operation of the copy request are synced to make sure the export is not unmounted while it's being used. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-22 03:09:37 +08:00
spin_unlock(&nn->nfsd_ssc_lock);
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
}
#else /* CONFIG_NFSD_V4_2_INTER_SSC */
static __be32
nfsd4_setup_inter_ssc(struct svc_rqst *rqstp,
struct nfsd4_compound_state *cstate,
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
struct nfsd4_copy *copy)
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
{
NFSD fix nfserro errno mismatch There is mismatch between __be32 and u32 in nfserr and errno. Reported-by: kbuild test robot <lkp@intel.com> Fixes: d5e54eeb0e3d ("NFSD add nfs4 inter ssc to nfsd4_copy") Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-12-05 04:13:53 +08:00
return nfserr_inval;
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
}
static void
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
nfsd4_cleanup_inter_ssc(struct nfsd4_ssc_umount_item *nsui, struct file *filp,
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
struct nfsd_file *dst)
{
}
static struct file *nfs42_ssc_open(struct vfsmount *ss_mnt,
struct nfs_fh *src_fh,
nfs4_stateid *stateid)
{
return NULL;
}
#endif /* CONFIG_NFSD_V4_2_INTER_SSC */
static __be32
nfsd4_setup_intra_ssc(struct svc_rqst *rqstp,
struct nfsd4_compound_state *cstate,
struct nfsd4_copy *copy)
{
return nfsd4_verify_copy(rqstp, cstate, &copy->cp_src_stateid,
&copy->nf_src, &copy->cp_dst_stateid,
&copy->nf_dst);
}
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
static void nfsd4_cb_offload_release(struct nfsd4_callback *cb)
{
NFSD: Move copy offload callback arguments into a separate structure Refactor so that CB_OFFLOAD arguments can be passed without allocating a whole struct nfsd4_copy object. On my system (x86_64) this removes another 96 bytes from struct nfsd4_copy. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:41:18 +08:00
struct nfsd4_cb_offload *cbo =
container_of(cb, struct nfsd4_cb_offload, co_cb);
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
NFSD: Move copy offload callback arguments into a separate structure Refactor so that CB_OFFLOAD arguments can be passed without allocating a whole struct nfsd4_copy object. On my system (x86_64) this removes another 96 bytes from struct nfsd4_copy. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:41:18 +08:00
kfree(cbo);
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
}
static int nfsd4_cb_offload_done(struct nfsd4_callback *cb,
struct rpc_task *task)
{
NFSD: Add tracepoints to report NFSv4 callback completions Wireshark has always been lousy about dissecting NFSv4 callbacks, especially NFSv4.0 backchannel requests. Add tracepoints so we can surgically capture these events in the trace log. Tracepoints are time-stamped and ordered so that we can now observe the timing relationship between a CB_RECALL Reply and the client's DELEGRETURN Call. Example: nfsd-1153 [002] 211.986391: nfsd_cb_recall: addr=192.168.1.67:45767 client 62ea82e4:fee7492a stateid 00000003:00000001 nfsd-1153 [002] 212.095634: nfsd_compound: xid=0x0000002c opcnt=2 nfsd-1153 [002] 212.095647: nfsd_compound_status: op=1/2 OP_PUTFH status=0 nfsd-1153 [002] 212.095658: nfsd_file_put: hash=0xf72 inode=0xffff9291148c7410 ref=3 flags=HASHED|REFERENCED may=READ file=0xffff929103b3ea00 nfsd-1153 [002] 212.095661: nfsd_compound_status: op=2/2 OP_DELEGRETURN status=0 kworker/u25:8-148 [002] 212.096713: nfsd_cb_recall_done: client 62ea82e4:fee7492a stateid 00000003:00000001 status=0 Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Reviewed-by: Jeff Layton <jlayton@kernel.org>
2022-09-09 06:13:54 +08:00
struct nfsd4_cb_offload *cbo =
container_of(cb, struct nfsd4_cb_offload, co_cb);
trace_nfsd_cb_offload_done(&cbo->co_res.cb_stateid, task);
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
return 1;
}
static const struct nfsd4_callback_ops nfsd4_cb_offload_ops = {
.release = nfsd4_cb_offload_release,
nfsd: track the main opcode for callbacks Keep track of the "main" opcode for the callback, and display it in the tracepoint. This makes it simpler to discern what's happening when there is more than one callback in flight. The one special case is the CB_NULL RPC. That's not a CB_COMPOUND opcode, so designate the value 0 for that. Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-08-26 20:50:12 +08:00
.done = nfsd4_cb_offload_done,
.opcode = OP_CB_OFFLOAD,
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
};
static void nfsd4_init_copy_res(struct nfsd4_copy *copy, bool sync)
{
NFSD add vfs_fsync after async copy is done Currently, the server does all copies as NFS_UNSTABLE. For synchronous copies linux client will append a COMMIT to the COPY compound but for async copies it does not (because COMMIT needs to be done after all bytes are copied and not as a reply to the COPY operation). However, in order to save the client doing a COMMIT as a separate rpc, the server can reply back with NFS_FILE_SYNC copy. This patch proposed to add vfs_fsync() call at the end of the async copy. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-20 02:48:27 +08:00
copy->cp_res.wr_stable_how =
NFSD: Replace boolean fields in struct nfsd4_copy Clean up: saves 8 bytes, and we can replace check_and_set_stop_copy() with an atomic bitop. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:41 +08:00
test_bit(NFSD4_COPY_F_COMMITTED, &copy->cp_flags) ?
NFS_FILE_SYNC : NFS_UNSTABLE;
nfsd4_copy_set_sync(copy, sync);
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
}
NFSD: Refactor nfsd4_do_copy() Refactor: Now that nfsd4_do_copy() no longer calls the cleanup helpers, plumb the use of struct file pointers all the way down to _nfsd_copy_file_range(). Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:59 +08:00
static ssize_t _nfsd_copy_file_range(struct nfsd4_copy *copy,
struct file *dst,
struct file *src)
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
{
nfsd: Replace use of rwsem with errseq_t The nfsd_file nf_rwsem is currently being used to separate file write and commit instances to ensure that we catch errors and apply them to the correct write/commit. We can improve scalability at the expense of a little accuracy (some extra false positives) by replacing the nf_rwsem with more careful use of the errseq_t mechanism to track errors across the different operations. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: rebased on zero-verifier fix ]
2021-12-19 09:38:01 +08:00
errseq_t since;
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
ssize_t bytes_copied = 0;
nfsd: don't ignore high bits of copy count Note size_t is 32-bit on a 32-bit architecture, but cp_count is defined by the protocol to be 64 bit, so we could be turning a large copy into a 0-length copy here. Reported-by: <radchenkoy@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2021-03-19 08:03:22 +08:00
u64 bytes_total = copy->cp_count;
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
u64 src_pos = copy->cp_src_pos;
u64 dst_pos = copy->cp_dst_pos;
NFSD: Fix sparse warning /home/cel/src/linux/linux/fs/nfsd/nfs4proc.c:1539:24: warning: incorrect type in assignment (different base types) /home/cel/src/linux/linux/fs/nfsd/nfs4proc.c:1539:24: expected restricted __be32 [usertype] status /home/cel/src/linux/linux/fs/nfsd/nfs4proc.c:1539:24: got int Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2021-10-14 04:44:20 +08:00
int status;
NFSD: pass range end to vfs_fsync_range() instead of count _nfsd_copy_file_range() calls vfs_fsync_range() with an offset and count (bytes written), but the former wants the start and end bytes of the range to sync. Fix it up. Fixes: eac0b17a77fb ("NFSD add vfs_fsync after async copy is done") Signed-off-by: Brian Foster <bfoster@redhat.com> Tested-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-11-16 23:28:36 +08:00
loff_t end;
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
nfsd: COPY with length 0 should copy to end of file >From https://tools.ietf.org/html/rfc7862#page-65 A count of 0 (zero) requests that all bytes from ca_src_offset through EOF be copied to the destination. Reported-by: <radchenkoy@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2021-03-19 08:03:23 +08:00
/* See RFC 7862 p.67: */
if (bytes_total == 0)
bytes_total = ULLONG_MAX;
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
do {
SUNRPC: change how svc threads are asked to exit. svc threads are currently stopped using kthread_stop(). This requires identifying a specific thread. However we don't care which thread stops, just as long as one does. So instead, set a flag in the svc_pool to say that a thread needs to die, and have each thread check this flag instead of calling kthread_should_stop(). The first thread to find and clear this flag then moves towards exiting. This removes an explicit dependency on sp_all_threads which will make a future patch simpler. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-09-11 22:39:04 +08:00
/* Only async copies can be stopped here */
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
if (kthread_should_stop())
break;
nfsd: Replace use of rwsem with errseq_t The nfsd_file nf_rwsem is currently being used to separate file write and commit instances to ensure that we catch errors and apply them to the correct write/commit. We can improve scalability at the expense of a little accuracy (some extra false positives) by replacing the nf_rwsem with more careful use of the errseq_t mechanism to track errors across the different operations. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: rebased on zero-verifier fix ]
2021-12-19 09:38:01 +08:00
bytes_copied = nfsd_copy_file_range(src, src_pos, dst, dst_pos,
bytes_total);
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
if (bytes_copied <= 0)
break;
bytes_total -= bytes_copied;
copy->cp_res.wr_bytes_written += bytes_copied;
src_pos += bytes_copied;
dst_pos += bytes_copied;
NFSD: Replace boolean fields in struct nfsd4_copy Clean up: saves 8 bytes, and we can replace check_and_set_stop_copy() with an atomic bitop. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:41 +08:00
} while (bytes_total > 0 && nfsd4_copy_is_async(copy));
NFSD add vfs_fsync after async copy is done Currently, the server does all copies as NFS_UNSTABLE. For synchronous copies linux client will append a COMMIT to the COPY compound but for async copies it does not (because COMMIT needs to be done after all bytes are copied and not as a reply to the COPY operation). However, in order to save the client doing a COMMIT as a separate rpc, the server can reply back with NFS_FILE_SYNC copy. This patch proposed to add vfs_fsync() call at the end of the async copy. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-20 02:48:27 +08:00
/* for a non-zero asynchronous copy do a commit of data */
NFSD: Replace boolean fields in struct nfsd4_copy Clean up: saves 8 bytes, and we can replace check_and_set_stop_copy() with an atomic bitop. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:41 +08:00
if (nfsd4_copy_is_async(copy) && copy->cp_res.wr_bytes_written > 0) {
nfsd: Replace use of rwsem with errseq_t The nfsd_file nf_rwsem is currently being used to separate file write and commit instances to ensure that we catch errors and apply them to the correct write/commit. We can improve scalability at the expense of a little accuracy (some extra false positives) by replacing the nf_rwsem with more careful use of the errseq_t mechanism to track errors across the different operations. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: rebased on zero-verifier fix ]
2021-12-19 09:38:01 +08:00
since = READ_ONCE(dst->f_wb_err);
NFSD: pass range end to vfs_fsync_range() instead of count _nfsd_copy_file_range() calls vfs_fsync_range() with an offset and count (bytes written), but the former wants the start and end bytes of the range to sync. Fix it up. Fixes: eac0b17a77fb ("NFSD add vfs_fsync after async copy is done") Signed-off-by: Brian Foster <bfoster@redhat.com> Tested-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-11-16 23:28:36 +08:00
end = copy->cp_dst_pos + copy->cp_res.wr_bytes_written - 1;
status = vfs_fsync_range(dst, copy->cp_dst_pos, end, 0);
nfsd: Replace use of rwsem with errseq_t The nfsd_file nf_rwsem is currently being used to separate file write and commit instances to ensure that we catch errors and apply them to the correct write/commit. We can improve scalability at the expense of a little accuracy (some extra false positives) by replacing the nf_rwsem with more careful use of the errseq_t mechanism to track errors across the different operations. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: rebased on zero-verifier fix ]
2021-12-19 09:38:01 +08:00
if (!status)
status = filemap_check_wb_err(dst->f_mapping, since);
NFSD add vfs_fsync after async copy is done Currently, the server does all copies as NFS_UNSTABLE. For synchronous copies linux client will append a COMMIT to the COPY compound but for async copies it does not (because COMMIT needs to be done after all bytes are copied and not as a reply to the COPY operation). However, in order to save the client doing a COMMIT as a separate rpc, the server can reply back with NFS_FILE_SYNC copy. This patch proposed to add vfs_fsync() call at the end of the async copy. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-20 02:48:27 +08:00
if (!status)
NFSD: Replace boolean fields in struct nfsd4_copy Clean up: saves 8 bytes, and we can replace check_and_set_stop_copy() with an atomic bitop. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:41 +08:00
set_bit(NFSD4_COPY_F_COMMITTED, &copy->cp_flags);
NFSD add vfs_fsync after async copy is done Currently, the server does all copies as NFS_UNSTABLE. For synchronous copies linux client will append a COMMIT to the COPY compound but for async copies it does not (because COMMIT needs to be done after all bytes are copied and not as a reply to the COPY operation). However, in order to save the client doing a COMMIT as a separate rpc, the server can reply back with NFS_FILE_SYNC copy. This patch proposed to add vfs_fsync() call at the end of the async copy. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-20 02:48:27 +08:00
}
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
return bytes_copied;
}
NFSD: Refactor nfsd4_do_copy() Refactor: Now that nfsd4_do_copy() no longer calls the cleanup helpers, plumb the use of struct file pointers all the way down to _nfsd_copy_file_range(). Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:59 +08:00
static __be32 nfsd4_do_copy(struct nfsd4_copy *copy,
struct file *src, struct file *dst,
bool sync)
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
{
__be32 status;
ssize_t bytes;
NFSD: Refactor nfsd4_do_copy() Refactor: Now that nfsd4_do_copy() no longer calls the cleanup helpers, plumb the use of struct file pointers all the way down to _nfsd_copy_file_range(). Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:59 +08:00
bytes = _nfsd_copy_file_range(copy, dst, src);
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
/* for async copy, we ignore the error, client can always retry
* to get the error
*/
if (bytes < 0 && !copy->cp_res.wr_bytes_written)
status = nfserrno(bytes);
else {
nfsd4_init_copy_res(copy, sync);
status = nfs_ok;
}
return status;
}
NFSD: Fix sparse warning in nfs4proc.c linux/fs/nfsd/nfs4proc.c:1542:24: warning: incorrect type in assignment (different base types) linux/fs/nfsd/nfs4proc.c:1542:24: expected restricted __be32 [assigned] [usertype] status linux/fs/nfsd/nfs4proc.c:1542:24: got int Clean-up: The dup_copy_fields() function returns only zero, so make it return void for now, and get rid of the return code check. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-12-01 05:17:40 +08:00
static void dup_copy_fields(struct nfsd4_copy *src, struct nfsd4_copy *dst)
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
{
dst->cp_src_pos = src->cp_src_pos;
dst->cp_dst_pos = src->cp_dst_pos;
dst->cp_count = src->cp_count;
NFSD: Replace boolean fields in struct nfsd4_copy Clean up: saves 8 bytes, and we can replace check_and_set_stop_copy() with an atomic bitop. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:41 +08:00
dst->cp_flags = src->cp_flags;
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
memcpy(&dst->cp_res, &src->cp_res, sizeof(src->cp_res));
memcpy(&dst->fh, &src->fh, sizeof(src->fh));
dst->cp_clp = src->cp_clp;
nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp. Since we now presume that the struct file will be persistent in most cases, we can stop fiddling with the raparms in the read code. This also means that we don't really care about the rd_tmp_file field anymore. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-08-19 02:18:54 +08:00
dst->nf_dst = nfsd_file_get(src->nf_dst);
NFSD: Replace boolean fields in struct nfsd4_copy Clean up: saves 8 bytes, and we can replace check_and_set_stop_copy() with an atomic bitop. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:41 +08:00
/* for inter, nf_src doesn't exist yet */
if (!nfsd4_ssc_is_inter(src))
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
dst->nf_src = nfsd_file_get(src->nf_src);
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
memcpy(&dst->cp_stateid, &src->cp_stateid, sizeof(src->cp_stateid));
NFSD: Shrink size of struct nfsd4_copy struct nfsd4_copy is part of struct nfsd4_op, which resides in an 8-element array. sizeof(struct nfsd4_op): Before: /* size: 1696, cachelines: 27, members: 5 */ After: /* size: 672, cachelines: 11, members: 5 */ Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:22 +08:00
memcpy(dst->cp_src, src->cp_src, sizeof(struct nl4_server));
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
memcpy(&dst->stateid, &src->stateid, sizeof(src->stateid));
memcpy(&dst->c_fh, &src->c_fh, sizeof(src->c_fh));
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
dst->ss_nsui = src->ss_nsui;
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
}
nfsd: clean up potential nfsd_file refcount leaks in COPY codepath There are two different flavors of the nfsd4_copy struct. One is embedded in the compound and is used directly in synchronous copies. The other is dynamically allocated, refcounted and tracked in the client struture. For the embedded one, the cleanup just involves releasing any nfsd_files held on its behalf. For the async one, the cleanup is a bit more involved, and we need to dequeue it from lists, unhash it, etc. There is at least one potential refcount leak in this code now. If the kthread_create call fails, then both the src and dst nfsd_files in the original nfsd4_copy object are leaked. The cleanup in this codepath is also sort of weird. In the async copy case, we'll have up to four nfsd_file references (src and dst for both flavors of copy structure). They are both put at the end of nfsd4_do_async_copy, even though the ones held on behalf of the embedded one outlive that structure. Change it so that we always clean up the nfsd_file refs held by the embedded copy structure before nfsd4_copy returns. Rework cleanup_async_copy to handle both inter and intra copies. Eliminate nfsd4_cleanup_intra_ssc since it now becomes a no-op. Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-01-18 03:38:31 +08:00
static void release_copy_files(struct nfsd4_copy *copy)
{
if (copy->nf_src)
nfsd_file_put(copy->nf_src);
if (copy->nf_dst)
nfsd_file_put(copy->nf_dst);
}
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
static void cleanup_async_copy(struct nfsd4_copy *copy)
{
NFSD add COPY_NOTIFY operation Introducing the COPY_NOTIFY operation. Create a new unique stateid that will keep track of the copy state and the upcoming READs that will use that stateid. Each associated parent stateid has a list of copy notify stateids. A copy notify structure makes a copy of the parent stateid and a clientid and will use it to look up the parent stateid during the READ request (suggested by Trond Myklebust <trond.myklebust@hammerspace.com>). At nfs4_put_stid() time, we walk the list of the associated copy notify stateids and delete them. Laundromat thread will traverse globally stored copy notify stateid in idr and notice if any haven't been referenced in the lease period, if so, it'll remove them. Return single netaddr to advertise to the copy. Suggested-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Andy Adamson <andros@netapp.com>
2019-10-05 04:34:26 +08:00
nfs4_free_copy_state(copy);
nfsd: clean up potential nfsd_file refcount leaks in COPY codepath There are two different flavors of the nfsd4_copy struct. One is embedded in the compound and is used directly in synchronous copies. The other is dynamically allocated, refcounted and tracked in the client struture. For the embedded one, the cleanup just involves releasing any nfsd_files held on its behalf. For the async one, the cleanup is a bit more involved, and we need to dequeue it from lists, unhash it, etc. There is at least one potential refcount leak in this code now. If the kthread_create call fails, then both the src and dst nfsd_files in the original nfsd4_copy object are leaked. The cleanup in this codepath is also sort of weird. In the async copy case, we'll have up to four nfsd_file references (src and dst for both flavors of copy structure). They are both put at the end of nfsd4_do_async_copy, even though the ones held on behalf of the embedded one outlive that structure. Change it so that we always clean up the nfsd_file refs held by the embedded copy structure before nfsd4_copy returns. Rework cleanup_async_copy to handle both inter and intra copies. Eliminate nfsd4_cleanup_intra_ssc since it now becomes a no-op. Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-01-18 03:38:31 +08:00
release_copy_files(copy);
NFSD: fix problems with cleanup on errors in nfsd4_copy When nfsd4_copy fails to allocate memory for async_copy->cp_src, or nfs4_init_copy_state fails, it calls cleanup_async_copy to do the cleanup for the async_copy which causes page fault since async_copy is not yet initialized. This patche rearranges the order of initializing the fields in async_copy and adds checks in cleanup_async_copy to skip un-initialized fields. Fixes: ce0887ac96d3 ("NFSD add nfs4 inter ssc to nfsd4_copy") Fixes: 87689df69491 ("NFSD: Shrink size of struct nfsd4_copy") Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-02-01 03:12:29 +08:00
if (copy->cp_clp) {
spin_lock(&copy->cp_clp->async_lock);
if (!list_empty(&copy->copies))
list_del_init(&copy->copies);
spin_unlock(&copy->cp_clp->async_lock);
}
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
nfs4_put_copy(copy);
}
NFSD: Record status of async copy operation in struct nfsd4_copy After a client has started an asynchronous COPY operation, a subsequent OFFLOAD_STATUS operation will need to report the status code once that COPY operation has completed. The recorded status record will be used by a subsequent patch. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-05-01 04:05:10 +08:00
static void nfsd4_send_cb_offload(struct nfsd4_copy *copy)
NFSD: Add nfsd4_send_cb_offload() Refactor for legibility. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:41:12 +08:00
{
NFSD: Move copy offload callback arguments into a separate structure Refactor so that CB_OFFLOAD arguments can be passed without allocating a whole struct nfsd4_copy object. On my system (x86_64) this removes another 96 bytes from struct nfsd4_copy. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:41:18 +08:00
struct nfsd4_cb_offload *cbo;
NFSD: Add nfsd4_send_cb_offload() Refactor for legibility. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:41:12 +08:00
NFSD: Move copy offload callback arguments into a separate structure Refactor so that CB_OFFLOAD arguments can be passed without allocating a whole struct nfsd4_copy object. On my system (x86_64) this removes another 96 bytes from struct nfsd4_copy. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:41:18 +08:00
cbo = kzalloc(sizeof(*cbo), GFP_KERNEL);
if (!cbo)
NFSD: Add nfsd4_send_cb_offload() Refactor for legibility. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:41:12 +08:00
return;
NFSD: Move copy offload callback arguments into a separate structure Refactor so that CB_OFFLOAD arguments can be passed without allocating a whole struct nfsd4_copy object. On my system (x86_64) this removes another 96 bytes from struct nfsd4_copy. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:41:18 +08:00
memcpy(&cbo->co_res, &copy->cp_res, sizeof(copy->cp_res));
memcpy(&cbo->co_fh, &copy->fh, sizeof(copy->fh));
NFSD: Record status of async copy operation in struct nfsd4_copy After a client has started an asynchronous COPY operation, a subsequent OFFLOAD_STATUS operation will need to report the status code once that COPY operation has completed. The recorded status record will be used by a subsequent patch. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-05-01 04:05:10 +08:00
cbo->co_nfserr = copy->nfserr;
NFSD: Add nfsd4_send_cb_offload() Refactor for legibility. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:41:12 +08:00
NFSD: Move copy offload callback arguments into a separate structure Refactor so that CB_OFFLOAD arguments can be passed without allocating a whole struct nfsd4_copy object. On my system (x86_64) this removes another 96 bytes from struct nfsd4_copy. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:41:18 +08:00
nfsd4_init_cb(&cbo->co_cb, copy->cp_clp, &nfsd4_cb_offload_ops,
NFSPROC4_CLNT_CB_OFFLOAD);
trace_nfsd_cb_offload(copy->cp_clp, &cbo->co_res.cb_stateid,
NFSD: Record status of async copy operation in struct nfsd4_copy After a client has started an asynchronous COPY operation, a subsequent OFFLOAD_STATUS operation will need to report the status code once that COPY operation has completed. The recorded status record will be used by a subsequent patch. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-05-01 04:05:10 +08:00
&cbo->co_fh, copy->cp_count, copy->nfserr);
NFSD: Move copy offload callback arguments into a separate structure Refactor so that CB_OFFLOAD arguments can be passed without allocating a whole struct nfsd4_copy object. On my system (x86_64) this removes another 96 bytes from struct nfsd4_copy. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:41:18 +08:00
nfsd4_run_cb(&cbo->co_cb);
NFSD: Add nfsd4_send_cb_offload() Refactor for legibility. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:41:12 +08:00
}
NFSD: Remove kmalloc from nfsd4_do_async_copy() Instead of manufacturing a phony struct nfsd_file, pass the struct file returned by nfs42_ssc_open() directly to nfsd4_do_copy(). Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:41:06 +08:00
/**
* nfsd4_do_async_copy - kthread function for background server-side COPY
* @data: arguments for COPY operation
*
* Return values:
* %0: Copy operation is done.
*/
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
static int nfsd4_do_async_copy(void *data)
{
struct nfsd4_copy *copy = (struct nfsd4_copy *)data;
NFSD: Wrap async copy operations with trace points Add an nfsd_copy_async_done to record the timestamp, the final status code, and the callback stateid of an async copy. Rename the nfsd_copy_do_async tracepoint to match that naming convention to make it easier to enable both of these with a single glob. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-08-29 01:40:09 +08:00
trace_nfsd_copy_async(copy);
NFSD: Replace boolean fields in struct nfsd4_copy Clean up: saves 8 bytes, and we can replace check_and_set_stop_copy() with an atomic bitop. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:41 +08:00
if (nfsd4_ssc_is_inter(copy)) {
NFSD: Remove kmalloc from nfsd4_do_async_copy() Instead of manufacturing a phony struct nfsd_file, pass the struct file returned by nfs42_ssc_open() directly to nfsd4_do_copy(). Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:41:06 +08:00
struct file *filp;
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
filp = nfs42_ssc_open(copy->ss_nsui->nsui_vfsmount,
&copy->c_fh, &copy->stateid);
NFSD: Remove kmalloc from nfsd4_do_async_copy() Instead of manufacturing a phony struct nfsd_file, pass the struct file returned by nfs42_ssc_open() directly to nfsd4_do_copy(). Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:41:06 +08:00
if (IS_ERR(filp)) {
NFSD enforce filehandle check for source file in COPY If the passed in filehandle for the source file in the COPY operation is not a regular file, the server MUST return NFS4ERR_WRONG_TYPE. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-08-20 03:16:36 +08:00
switch (PTR_ERR(filp)) {
case -EBADF:
NFSD: Record status of async copy operation in struct nfsd4_copy After a client has started an asynchronous COPY operation, a subsequent OFFLOAD_STATUS operation will need to report the status code once that COPY operation has completed. The recorded status record will be used by a subsequent patch. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-05-01 04:05:10 +08:00
copy->nfserr = nfserr_wrong_type;
NFSD enforce filehandle check for source file in COPY If the passed in filehandle for the source file in the COPY operation is not a regular file, the server MUST return NFS4ERR_WRONG_TYPE. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-08-20 03:16:36 +08:00
break;
default:
NFSD: Record status of async copy operation in struct nfsd4_copy After a client has started an asynchronous COPY operation, a subsequent OFFLOAD_STATUS operation will need to report the status code once that COPY operation has completed. The recorded status record will be used by a subsequent patch. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-05-01 04:05:10 +08:00
copy->nfserr = nfserr_offload_denied;
NFSD enforce filehandle check for source file in COPY If the passed in filehandle for the source file in the COPY operation is not a regular file, the server MUST return NFS4ERR_WRONG_TYPE. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-08-20 03:16:36 +08:00
}
NFSD: fix use-after-free in __nfs42_ssc_open() Problem caused by source's vfsmount being unmounted but remains on the delayed unmount list. This happens when nfs42_ssc_open() return errors. Fixed by removing nfsd4_interssc_connect(), leave the vfsmount for the laundromat to unmount when idle time expires. We don't need to call nfs_do_sb_deactive when nfs42_ssc_open return errors since the file was not opened so nfs_server->active was not incremented. Same as in nfsd4_copy, if we fail to launch nfsd4_do_async_copy thread then there's no need to call nfs_do_sb_deactive Reported-by: Xingyuan Mo <hdthky0@gmail.com> Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Tested-by: Xingyuan Mo <hdthky0@gmail.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-12-13 06:50:11 +08:00
/* ss_mnt will be unmounted by the laundromat */
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
goto do_callback;
}
NFSD: Record status of async copy operation in struct nfsd4_copy After a client has started an asynchronous COPY operation, a subsequent OFFLOAD_STATUS operation will need to report the status code once that COPY operation has completed. The recorded status record will be used by a subsequent patch. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-05-01 04:05:10 +08:00
copy->nfserr = nfsd4_do_copy(copy, filp, copy->nf_dst->nf_file,
false);
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
nfsd4_cleanup_inter_ssc(copy->ss_nsui, filp, copy->nf_dst);
NFSD: Refactor nfsd4_cleanup_inter_ssc() (2/2) Move the nfsd4_cleanup_*() call sites out of nfsd4_do_copy(). A subsequent patch will modify one of the new call sites to avoid the need to manufacture the phony struct nfsd_file. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:53 +08:00
} else {
NFSD: Record status of async copy operation in struct nfsd4_copy After a client has started an asynchronous COPY operation, a subsequent OFFLOAD_STATUS operation will need to report the status code once that COPY operation has completed. The recorded status record will be used by a subsequent patch. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-05-01 04:05:10 +08:00
copy->nfserr = nfsd4_do_copy(copy, copy->nf_src->nf_file,
copy->nf_dst->nf_file, false);
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
}
do_callback:
NFSD: Add COPY status code to OFFLOAD_STATUS response Clients that send an OFFLOAD_STATUS might want to distinguish between an async COPY operation that is still running, has completed successfully, or that has failed. The intention of this patch is to make NFSD behave like this: * Copy still running: OFFLOAD_STATUS returns NFS4_OK, the number of bytes copied so far, and an empty osr_status array * Copy completed successfully: OFFLOAD_STATUS returns NFS4_OK, the number of bytes copied, and an osr_status of NFS4_OK * Copy failed: OFFLOAD_STATUS returns NFS4_OK, the number of bytes copied, and an osr_status other than NFS4_OK * Copy operation lost, canceled, or otherwise unrecognized: OFFLOAD_STATUS returns NFS4ERR_BAD_STATEID NB: Though RFC 7862 Section 11.2 lists a small set of NFS status codes that are valid for OFFLOAD_STATUS, there do not seem to be any explicit spec limits on the status codes that may be returned in the osr_status field. At this time we have no unit tests for COPY and its brethren, as pynfs does not yet implement support for NFSv4.2. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-05-01 04:05:11 +08:00
set_bit(NFSD4_COPY_F_COMPLETED, &copy->cp_flags);
NFSD: Wrap async copy operations with trace points Add an nfsd_copy_async_done to record the timestamp, the final status code, and the callback stateid of an async copy. Rename the nfsd_copy_do_async tracepoint to match that naming convention to make it easier to enable both of these with a single glob. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-08-29 01:40:09 +08:00
trace_nfsd_copy_async_done(copy);
NFSD: Record status of async copy operation in struct nfsd4_copy After a client has started an asynchronous COPY operation, a subsequent OFFLOAD_STATUS operation will need to report the status code once that COPY operation has completed. The recorded status record will be used by a subsequent patch. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-05-01 04:05:10 +08:00
nfsd4_send_cb_offload(copy);
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
cleanup_async_copy(copy);
return 0;
}
NFSD: Implement the COPY call I only implemented the sync version of this call, since it's the easiest. I can simply call vfs_copy_range() and have the vfs do the right thing for the filesystem being exported. Signed-off-by: Anna Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-09-08 03:57:30 +08:00
static __be32
nfsd4_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
NFSD: Implement the COPY call I only implemented the sync version of this call, since it's the easiest. I can simply call vfs_copy_range() and have the vfs do the right thing for the filesystem being exported. Signed-off-by: Anna Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-09-08 03:57:30 +08:00
{
NFSD: Async COPY result needs to return a write verifier Currently, when NFSD handles an asynchronous COPY, it returns a zero write verifier, relying on the subsequent CB_OFFLOAD callback to pass the write verifier and a stable_how4 value to the client. However, if the CB_OFFLOAD never arrives at the client (for example, if a network partition occurs just as the server sends the CB_OFFLOAD operation), the client will never receive this verifier. Thus, if the client sends a follow-up COMMIT, there is no way for the client to assess the COMMIT result. The usual recovery for a missing CB_OFFLOAD is for the client to send an OFFLOAD_STATUS operation, but that operation does not carry a write verifier in its result. Neither does it carry a stable_how4 value, so the client /must/ send a COMMIT in this case -- which will always fail because currently there's still no write verifier in the COPY result. Thus the server needs to return a normal write verifier in its COPY result even if the COPY operation is to be performed asynchronously. If the server recognizes the callback stateid in subsequent OFFLOAD_STATUS operations, then obviously it has not restarted, and the write verifier the client received in the COPY result is still valid and can be used to assess a COMMIT of the copied data, if one is needed. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-08-29 01:40:03 +08:00
struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
struct nfsd4_copy *async_copy = NULL;
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_copy *copy = &u->copy;
NFSD: Async COPY result needs to return a write verifier Currently, when NFSD handles an asynchronous COPY, it returns a zero write verifier, relying on the subsequent CB_OFFLOAD callback to pass the write verifier and a stable_how4 value to the client. However, if the CB_OFFLOAD never arrives at the client (for example, if a network partition occurs just as the server sends the CB_OFFLOAD operation), the client will never receive this verifier. Thus, if the client sends a follow-up COMMIT, there is no way for the client to assess the COMMIT result. The usual recovery for a missing CB_OFFLOAD is for the client to send an OFFLOAD_STATUS operation, but that operation does not carry a write verifier in its result. Neither does it carry a stable_how4 value, so the client /must/ send a COMMIT in this case -- which will always fail because currently there's still no write verifier in the COPY result. Thus the server needs to return a normal write verifier in its COPY result even if the COPY operation is to be performed asynchronously. If the server recognizes the callback stateid in subsequent OFFLOAD_STATUS operations, then obviously it has not restarted, and the write verifier the client received in the COPY result is still valid and can be used to assess a COMMIT of the copied data, if one is needed. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-08-29 01:40:03 +08:00
struct nfsd42_write_res *result;
NFSD: Implement the COPY call I only implemented the sync version of this call, since it's the easiest. I can simply call vfs_copy_range() and have the vfs do the right thing for the filesystem being exported. Signed-off-by: Anna Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-09-08 03:57:30 +08:00
__be32 status;
NFSD: Force all NFSv4.2 COPY requests to be synchronous We've discovered that delivering a CB_OFFLOAD operation can be unreliable in some pretty unremarkable situations. Examples include: - The server dropped the connection because it lost a forechannel NFSv4 request and wishes to force the client to retransmit - The GSS sequence number window under-flowed - A network partition occurred When that happens, all pending callback operations, including CB_OFFLOAD, are lost. NFSD does not retransmit them. Moreover, the Linux NFS client does not yet support sending an OFFLOAD_STATUS operation to probe whether an asynchronous COPY operation has finished. Thus, on Linux NFS clients, when a CB_OFFLOAD is lost, asynchronous COPY can hang until manually interrupted. I've tried a couple of remedies, but so far the side-effects are worse than the disease and they have had to be reverted. So temporarily force COPY operations to be synchronous so that the use of CB_OFFLOAD is avoided entirely. This is a fix that can easily be backported to LTS kernels. I am working on client patches that introduce an implementation of OFFLOAD_STATUS. Note that NFSD arbitrarily limits the size of a copy_file_range to 4MB to avoid indefinitely blocking an nfsd thread. A short COPY result is returned in that case, and the client can present a fresh COPY request for the remainder. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-05-07 21:37:14 +08:00
/*
* Currently, async COPY is not reliable. Force all COPY
* requests to be synchronous to avoid client application
* hangs waiting for COPY completion.
*/
nfsd4_copy_set_sync(copy, true);
NFSD: Async COPY result needs to return a write verifier Currently, when NFSD handles an asynchronous COPY, it returns a zero write verifier, relying on the subsequent CB_OFFLOAD callback to pass the write verifier and a stable_how4 value to the client. However, if the CB_OFFLOAD never arrives at the client (for example, if a network partition occurs just as the server sends the CB_OFFLOAD operation), the client will never receive this verifier. Thus, if the client sends a follow-up COMMIT, there is no way for the client to assess the COMMIT result. The usual recovery for a missing CB_OFFLOAD is for the client to send an OFFLOAD_STATUS operation, but that operation does not carry a write verifier in its result. Neither does it carry a stable_how4 value, so the client /must/ send a COMMIT in this case -- which will always fail because currently there's still no write verifier in the COPY result. Thus the server needs to return a normal write verifier in its COPY result even if the COPY operation is to be performed asynchronously. If the server recognizes the callback stateid in subsequent OFFLOAD_STATUS operations, then obviously it has not restarted, and the write verifier the client received in the COPY result is still valid and can be used to assess a COMMIT of the copied data, if one is needed. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-08-29 01:40:03 +08:00
result = &copy->cp_res;
nfsd_copy_write_verifier((__be32 *)&result->wr_verifier.data, nn);
NFSD: initialize copy->cp_clp early in nfsd4_copy for use by trace point Prepare for adding server copy trace points. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Tested-by: Chen Hanxiao <chenhx.fnst@fujitsu.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-08-31 07:46:58 +08:00
copy->cp_clp = cstate->clp;
NFSD: Replace boolean fields in struct nfsd4_copy Clean up: saves 8 bytes, and we can replace check_and_set_stop_copy() with an atomic bitop. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:41 +08:00
if (nfsd4_ssc_is_inter(copy)) {
NFSD: add trace points to track server copy progress Add trace points on destination server to track inter and intra server copy operations. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Tested-by: Chen Hanxiao <chenhx.fnst@fujitsu.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-09-01 03:35:47 +08:00
trace_nfsd_copy_inter(copy);
NFSD: Replace boolean fields in struct nfsd4_copy Clean up: saves 8 bytes, and we can replace check_and_set_stop_copy() with an atomic bitop. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:41 +08:00
if (!inter_copy_offload_enable || nfsd4_copy_is_sync(copy)) {
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
status = nfserr_notsupp;
goto out;
}
NFSD: enhance inter-server copy cleanup Currently nfsd4_setup_inter_ssc returns the vfsmount of the source server's export when the mount completes. After the copy is done nfsd4_cleanup_inter_ssc is called with the vfsmount of the source server and it searches nfsd_ssc_mount_list for a matching entry to do the clean up. The problems with this approach are (1) the need to search the nfsd_ssc_mount_list and (2) the code has to handle the case where the matching entry is not found which looks ugly. The enhancement is instead of nfsd4_setup_inter_ssc returning the vfsmount, it returns the nfsd4_ssc_umount_item which has the vfsmount embedded in it. When nfsd4_cleanup_inter_ssc is called it's passed with the nfsd4_ssc_umount_item directly to do the clean up so no searching is needed and there is no need to handle the 'not found' case. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [ cel: adjusted whitespace and variable/function names ] Reviewed-by: Olga Kornievskaia <kolga@netapp.com>
2022-12-19 08:55:53 +08:00
status = nfsd4_setup_inter_ssc(rqstp, cstate, copy);
NFSD: add trace points to track server copy progress Add trace points on destination server to track inter and intra server copy operations. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Tested-by: Chen Hanxiao <chenhx.fnst@fujitsu.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-09-01 03:35:47 +08:00
if (status) {
trace_nfsd_copy_done(copy, status);
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
return nfserr_offload_denied;
NFSD: add trace points to track server copy progress Add trace points on destination server to track inter and intra server copy operations. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Tested-by: Chen Hanxiao <chenhx.fnst@fujitsu.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-09-01 03:35:47 +08:00
}
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
} else {
NFSD: add trace points to track server copy progress Add trace points on destination server to track inter and intra server copy operations. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Tested-by: Chen Hanxiao <chenhx.fnst@fujitsu.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-09-01 03:35:47 +08:00
trace_nfsd_copy_intra(copy);
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
status = nfsd4_setup_intra_ssc(rqstp, cstate, copy);
NFSD: add trace points to track server copy progress Add trace points on destination server to track inter and intra server copy operations. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Tested-by: Chen Hanxiao <chenhx.fnst@fujitsu.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-09-01 03:35:47 +08:00
if (status) {
trace_nfsd_copy_done(copy, status);
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
return status;
NFSD: add trace points to track server copy progress Add trace points on destination server to track inter and intra server copy operations. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Tested-by: Chen Hanxiao <chenhx.fnst@fujitsu.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-09-01 03:35:47 +08:00
}
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
}
NFSD: Implement the COPY call I only implemented the sync version of this call, since it's the easiest. I can simply call vfs_copy_range() and have the vfs do the right thing for the filesystem being exported. Signed-off-by: Anna Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-09-08 03:57:30 +08:00
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
memcpy(&copy->fh, &cstate->current_fh.fh_handle,
sizeof(struct knfsd_fh));
NFSD: Replace boolean fields in struct nfsd4_copy Clean up: saves 8 bytes, and we can replace check_and_set_stop_copy() with an atomic bitop. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:41 +08:00
if (nfsd4_copy_is_async(copy)) {
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
async_copy = kzalloc(sizeof(struct nfsd4_copy), GFP_KERNEL);
if (!async_copy)
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
goto out_err;
NFSD: Limit the number of concurrent async COPY operations Nothing appears to limit the number of concurrent async COPY operations that clients can start. In addition, AFAICT each async COPY can copy an unlimited number of 4MB chunks, so can run for a long time. Thus IMO async COPY can become a DoS vector. Add a restriction mechanism that bounds the number of concurrent background COPY operations. Start simple and try to be fair -- this patch implements a per-namespace limit. An async COPY request that occurs while this limit is exceeded gets NFS4ERR_DELAY. The requesting client can choose to send the request again after a delay or fall back to a traditional read/write style copy. If there is need to make the mechanism more sophisticated, we can visit that in future patches. Cc: stable@vger.kernel.org Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-08-29 01:40:04 +08:00
async_copy->cp_nn = nn;
NFSD: Initialize struct nfsd4_copy earlier Ensure the refcount and async_copies fields are initialized early. cleanup_async_copy() will reference these fields if an error occurs in nfsd4_copy(). If they are not correctly initialized, at the very least, a refcount underflow occurs. Reported-by: Olga Kornievskaia <okorniev@redhat.com> Fixes: aadc3bbea163 ("NFSD: Limit the number of concurrent async COPY operations") Reviewed-by: Jeff Layton <jlayton@kernel.org> Tested-by: Olga Kornievskaia <okorniev@redhat.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-10-27 00:02:38 +08:00
INIT_LIST_HEAD(&async_copy->copies);
refcount_set(&async_copy->refcount, 1);
NFSD: Limit the number of concurrent async COPY operations Nothing appears to limit the number of concurrent async COPY operations that clients can start. In addition, AFAICT each async COPY can copy an unlimited number of 4MB chunks, so can run for a long time. Thus IMO async COPY can become a DoS vector. Add a restriction mechanism that bounds the number of concurrent background COPY operations. Start simple and try to be fair -- this patch implements a per-namespace limit. An async COPY request that occurs while this limit is exceeded gets NFS4ERR_DELAY. The requesting client can choose to send the request again after a delay or fall back to a traditional read/write style copy. If there is need to make the mechanism more sophisticated, we can visit that in future patches. Cc: stable@vger.kernel.org Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-08-29 01:40:04 +08:00
/* Arbitrary cap on number of pending async copy operations */
if (atomic_inc_return(&nn->pending_async_copies) >
NFSD: Never decrement pending_async_copies on error The error flow in nfsd4_copy() calls cleanup_async_copy(), which already decrements nn->pending_async_copies. Reported-by: Olga Kornievskaia <okorniev@redhat.com> Fixes: aadc3bbea163 ("NFSD: Limit the number of concurrent async COPY operations") Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-10-30 03:27:19 +08:00
(int)rqstp->rq_pool->sp_nrthreads)
NFSD: Limit the number of concurrent async COPY operations Nothing appears to limit the number of concurrent async COPY operations that clients can start. In addition, AFAICT each async COPY can copy an unlimited number of 4MB chunks, so can run for a long time. Thus IMO async COPY can become a DoS vector. Add a restriction mechanism that bounds the number of concurrent background COPY operations. Start simple and try to be fair -- this patch implements a per-namespace limit. An async COPY request that occurs while this limit is exceeded gets NFS4ERR_DELAY. The requesting client can choose to send the request again after a delay or fall back to a traditional read/write style copy. If there is need to make the mechanism more sophisticated, we can visit that in future patches. Cc: stable@vger.kernel.org Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-08-29 01:40:04 +08:00
goto out_err;
NFSD: Shrink size of struct nfsd4_copy struct nfsd4_copy is part of struct nfsd4_op, which resides in an 8-element array. sizeof(struct nfsd4_op): Before: /* size: 1696, cachelines: 27, members: 5 */ After: /* size: 672, cachelines: 11, members: 5 */ Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:22 +08:00
async_copy->cp_src = kmalloc(sizeof(*async_copy->cp_src), GFP_KERNEL);
if (!async_copy->cp_src)
goto out_err;
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
if (!nfs4_init_copy_state(nn, copy))
goto out_err;
NFSD: Async COPY result needs to return a write verifier Currently, when NFSD handles an asynchronous COPY, it returns a zero write verifier, relying on the subsequent CB_OFFLOAD callback to pass the write verifier and a stable_how4 value to the client. However, if the CB_OFFLOAD never arrives at the client (for example, if a network partition occurs just as the server sends the CB_OFFLOAD operation), the client will never receive this verifier. Thus, if the client sends a follow-up COMMIT, there is no way for the client to assess the COMMIT result. The usual recovery for a missing CB_OFFLOAD is for the client to send an OFFLOAD_STATUS operation, but that operation does not carry a write verifier in its result. Neither does it carry a stable_how4 value, so the client /must/ send a COMMIT in this case -- which will always fail because currently there's still no write verifier in the COPY result. Thus the server needs to return a normal write verifier in its COPY result even if the COPY operation is to be performed asynchronously. If the server recognizes the callback stateid in subsequent OFFLOAD_STATUS operations, then obviously it has not restarted, and the write verifier the client received in the COPY result is still valid and can be used to assess a COMMIT of the copied data, if one is needed. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-08-29 01:40:03 +08:00
memcpy(&result->cb_stateid, &copy->cp_stateid.cs_stid,
sizeof(result->cb_stateid));
NFSD: Fix sparse warning in nfs4proc.c linux/fs/nfsd/nfs4proc.c:1542:24: warning: incorrect type in assignment (different base types) linux/fs/nfsd/nfs4proc.c:1542:24: expected restricted __be32 [assigned] [usertype] status linux/fs/nfsd/nfs4proc.c:1542:24: got int Clean-up: The dup_copy_fields() function returns only zero, so make it return void for now, and get rid of the return code check. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-12-01 05:17:40 +08:00
dup_copy_fields(copy, async_copy);
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
async_copy->copy_task = kthread_create(nfsd4_do_async_copy,
async_copy, "%s", "copy thread");
if (IS_ERR(async_copy->copy_task))
goto out_err;
spin_lock(&async_copy->cp_clp->async_lock);
list_add(&async_copy->copies,
&async_copy->cp_clp->async_copies);
spin_unlock(&async_copy->cp_clp->async_lock);
wake_up_process(async_copy->copy_task);
NFSD: Implement the COPY call I only implemented the sync version of this call, since it's the easiest. I can simply call vfs_copy_range() and have the vfs do the right thing for the filesystem being exported. Signed-off-by: Anna Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-09-08 03:57:30 +08:00
status = nfs_ok;
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
} else {
NFSD: Refactor nfsd4_do_copy() Refactor: Now that nfsd4_do_copy() no longer calls the cleanup helpers, plumb the use of struct file pointers all the way down to _nfsd_copy_file_range(). Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:59 +08:00
status = nfsd4_do_copy(copy, copy->nf_src->nf_file,
copy->nf_dst->nf_file, true);
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
}
NFSD: Implement the COPY call I only implemented the sync version of this call, since it's the easiest. I can simply call vfs_copy_range() and have the vfs do the right thing for the filesystem being exported. Signed-off-by: Anna Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-09-08 03:57:30 +08:00
out:
NFSD: add trace points to track server copy progress Add trace points on destination server to track inter and intra server copy operations. Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Tested-by: Chen Hanxiao <chenhx.fnst@fujitsu.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-09-01 03:35:47 +08:00
trace_nfsd_copy_done(copy, status);
nfsd: clean up potential nfsd_file refcount leaks in COPY codepath There are two different flavors of the nfsd4_copy struct. One is embedded in the compound and is used directly in synchronous copies. The other is dynamically allocated, refcounted and tracked in the client struture. For the embedded one, the cleanup just involves releasing any nfsd_files held on its behalf. For the async one, the cleanup is a bit more involved, and we need to dequeue it from lists, unhash it, etc. There is at least one potential refcount leak in this code now. If the kthread_create call fails, then both the src and dst nfsd_files in the original nfsd4_copy object are leaked. The cleanup in this codepath is also sort of weird. In the async copy case, we'll have up to four nfsd_file references (src and dst for both flavors of copy structure). They are both put at the end of nfsd4_do_async_copy, even though the ones held on behalf of the embedded one outlive that structure. Change it so that we always clean up the nfsd_file refs held by the embedded copy structure before nfsd4_copy returns. Rework cleanup_async_copy to handle both inter and intra copies. Eliminate nfsd4_cleanup_intra_ssc since it now becomes a no-op. Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-01-18 03:38:31 +08:00
release_copy_files(copy);
NFSD: Implement the COPY call I only implemented the sync version of this call, since it's the easiest. I can simply call vfs_copy_range() and have the vfs do the right thing for the filesystem being exported. Signed-off-by: Anna Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-09-08 03:57:30 +08:00
return status;
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
out_err:
NFSD: fix leaked reference count of nfsd4_ssc_umount_item The reference count of nfsd4_ssc_umount_item is not decremented on error conditions. This prevents the laundromat from unmounting the vfsmount of the source file. This patch decrements the reference count of nfsd4_ssc_umount_item on error. Fixes: f4e44b393389 ("NFSD: delay unmount source's export after inter-server copy completed.") Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-01-24 13:34:13 +08:00
if (nfsd4_ssc_is_inter(copy)) {
/*
* Source's vfsmount of inter-copy will be unmounted
* by the laundromat. Use copy instead of async_copy
* since async_copy->ss_nsui might not be set yet.
*/
refcount_dec(&copy->ss_nsui->nsui_refcnt);
}
NFSD fixing possible null pointer derefering in copy offload Static checker revealed possible error path leading to possible NULL pointer dereferencing. Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Fixes: e0639dc5805a: ("NFSD introduce async copy feature") Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-12-05 04:13:54 +08:00
if (async_copy)
cleanup_async_copy(async_copy);
NFSD: Limit the number of concurrent async COPY operations Nothing appears to limit the number of concurrent async COPY operations that clients can start. In addition, AFAICT each async COPY can copy an unlimited number of 4MB chunks, so can run for a long time. Thus IMO async COPY can become a DoS vector. Add a restriction mechanism that bounds the number of concurrent background COPY operations. Start simple and try to be fair -- this patch implements a per-namespace limit. An async COPY request that occurs while this limit is exceeded gets NFS4ERR_DELAY. The requesting client can choose to send the request again after a delay or fall back to a traditional read/write style copy. If there is need to make the mechanism more sophisticated, we can visit that in future patches. Cc: stable@vger.kernel.org Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-08-29 01:40:04 +08:00
status = nfserr_jukebox;
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
goto out;
}
nfsd: don't take nfsd4_copy ref for OP_OFFLOAD_STATUS We're not doing any blocking operations for OP_OFFLOAD_STATUS, so taking and putting a reference is a waste of effort. Take the client lock, search for the copy and fetch the wr_bytes_written field and return. Also, make find_async_copy a static function. Signed-off-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-01-19 01:31:34 +08:00
static struct nfsd4_copy *
find_async_copy_locked(struct nfs4_client *clp, stateid_t *stateid)
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
{
struct nfsd4_copy *copy;
nfsd: don't take nfsd4_copy ref for OP_OFFLOAD_STATUS We're not doing any blocking operations for OP_OFFLOAD_STATUS, so taking and putting a reference is a waste of effort. Take the client lock, search for the copy and fetch the wr_bytes_written field and return. Also, make find_async_copy a static function. Signed-off-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-01-19 01:31:34 +08:00
lockdep_assert_held(&clp->async_lock);
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
list_for_each_entry(copy, &clp->async_copies, copies) {
NFSD: Rename the fields in copy_stateid_t Code maintenance: The name of the copy_stateid_t::sc_count field collides with the sc_count field in struct nfs4_stid, making the latter difficult to grep for when auditing stateid reference counting. No behavior change expected. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-23 01:10:35 +08:00
if (memcmp(&copy->cp_stateid.cs_stid, stateid, NFS4_STATEID_SIZE))
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
continue;
return copy;
}
return NULL;
NFSD: Implement the COPY call I only implemented the sync version of this call, since it's the easiest. I can simply call vfs_copy_range() and have the vfs do the right thing for the filesystem being exported. Signed-off-by: Anna Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-09-08 03:57:30 +08:00
}
nfsd: don't take nfsd4_copy ref for OP_OFFLOAD_STATUS We're not doing any blocking operations for OP_OFFLOAD_STATUS, so taking and putting a reference is a waste of effort. Take the client lock, search for the copy and fetch the wr_bytes_written field and return. Also, make find_async_copy a static function. Signed-off-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-01-19 01:31:34 +08:00
static struct nfsd4_copy *
find_async_copy(struct nfs4_client *clp, stateid_t *stateid)
{
struct nfsd4_copy *copy;
spin_lock(&clp->async_lock);
copy = find_async_copy_locked(clp, stateid);
if (copy)
refcount_inc(&copy->refcount);
spin_unlock(&clp->async_lock);
return copy;
}
NFSD OFFLOAD_CANCEL xdr Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:19 +08:00
static __be32
nfsd4_offload_cancel(struct svc_rqst *rqstp,
struct nfsd4_compound_state *cstate,
union nfsd4_op_u *u)
{
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
struct nfsd4_offload_status *os = &u->offload_status;
struct nfsd4_copy *copy;
struct nfs4_client *clp = cstate->clp;
copy = find_async_copy(clp, &os->stateid);
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
if (!copy) {
struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
return manage_cpntf_state(nn, &os->stateid, clp, NULL);
} else
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
nfsd4_stop_copy(copy);
NFSD add nfs4 inter ssc to nfsd4_copy Given a universal address, mount the source server from the destination server. Use an internal mount. Call the NFS client nfs42_ssc_open to obtain the NFS struct file suitable for nfsd_copy_range. Ability to do "inter" server-to-server depends on the an nfsd kernel parameter "inter_copy_offload_enable". Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-09 23:50:48 +08:00
return nfs_ok;
NFSD OFFLOAD_CANCEL xdr Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:19 +08:00
}
NFSD COPY_NOTIFY xdr Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-08-08 23:14:59 +08:00
static __be32
nfsd4_copy_notify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
union nfsd4_op_u *u)
{
NFSD add COPY_NOTIFY operation Introducing the COPY_NOTIFY operation. Create a new unique stateid that will keep track of the copy state and the upcoming READs that will use that stateid. Each associated parent stateid has a list of copy notify stateids. A copy notify structure makes a copy of the parent stateid and a clientid and will use it to look up the parent stateid during the READ request (suggested by Trond Myklebust <trond.myklebust@hammerspace.com>). At nfs4_put_stid() time, we walk the list of the associated copy notify stateids and delete them. Laundromat thread will traverse globally stored copy notify stateid in idr and notice if any haven't been referenced in the lease period, if so, it'll remove them. Return single netaddr to advertise to the copy. Suggested-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Andy Adamson <andros@netapp.com>
2019-10-05 04:34:26 +08:00
struct nfsd4_copy_notify *cn = &u->copy_notify;
__be32 status;
struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
nfsd: don't assume copy notify when preprocessing the stateid Move the stateid handling to nfsd4_copy_notify. If nfs4_preprocess_stateid_op did not produce an output stateid, error out. Copy notify specifically does not permit the use of special stateids, so enforce that outside generic stateid pre-processing. Signed-off-by: Sagi Grimberg <sagi@grimberg.me> Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Olga Kornievskaia <aglo@umich.edu> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-07-25 01:01:37 +08:00
struct nfs4_stid *stid = NULL;
NFSD add COPY_NOTIFY operation Introducing the COPY_NOTIFY operation. Create a new unique stateid that will keep track of the copy state and the upcoming READs that will use that stateid. Each associated parent stateid has a list of copy notify stateids. A copy notify structure makes a copy of the parent stateid and a clientid and will use it to look up the parent stateid during the READ request (suggested by Trond Myklebust <trond.myklebust@hammerspace.com>). At nfs4_put_stid() time, we walk the list of the associated copy notify stateids and delete them. Laundromat thread will traverse globally stored copy notify stateid in idr and notice if any haven't been referenced in the lease period, if so, it'll remove them. Return single netaddr to advertise to the copy. Suggested-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Andy Adamson <andros@netapp.com>
2019-10-05 04:34:26 +08:00
struct nfs4_cpntf_state *cps;
struct nfs4_client *clp = cstate->clp;
status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
&cn->cpn_src_stateid, RD_STATE, NULL,
&stid);
if (status)
return status;
nfsd: don't assume copy notify when preprocessing the stateid Move the stateid handling to nfsd4_copy_notify. If nfs4_preprocess_stateid_op did not produce an output stateid, error out. Copy notify specifically does not permit the use of special stateids, so enforce that outside generic stateid pre-processing. Signed-off-by: Sagi Grimberg <sagi@grimberg.me> Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Olga Kornievskaia <aglo@umich.edu> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-07-25 01:01:37 +08:00
if (!stid)
return nfserr_bad_stateid;
NFSD add COPY_NOTIFY operation Introducing the COPY_NOTIFY operation. Create a new unique stateid that will keep track of the copy state and the upcoming READs that will use that stateid. Each associated parent stateid has a list of copy notify stateids. A copy notify structure makes a copy of the parent stateid and a clientid and will use it to look up the parent stateid during the READ request (suggested by Trond Myklebust <trond.myklebust@hammerspace.com>). At nfs4_put_stid() time, we walk the list of the associated copy notify stateids and delete them. Laundromat thread will traverse globally stored copy notify stateid in idr and notice if any haven't been referenced in the lease period, if so, it'll remove them. Return single netaddr to advertise to the copy. Suggested-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Andy Adamson <andros@netapp.com>
2019-10-05 04:34:26 +08:00
NFSD: Clean up nfsd4_encode_copy_notify() Replace open-coded encoding logic with the use of conventional XDR utility functions. Note that if we replace the cpn_sec and cpn_nsec fields with a single struct timespec64 field, the encoder can use nfsd4_encode_nfstime4(), as that is the data type specified by the XDR spec. NFS4ERR_INVAL seems inappropriate if the encoder doesn't support encoding the response. Instead use NFS4ERR_SERVERFAULT, since this condition is a software bug on the server. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-10-10 02:30:16 +08:00
cn->cpn_lease_time.tv_sec = nn->nfsd4_lease;
cn->cpn_lease_time.tv_nsec = 0;
NFSD add COPY_NOTIFY operation Introducing the COPY_NOTIFY operation. Create a new unique stateid that will keep track of the copy state and the upcoming READs that will use that stateid. Each associated parent stateid has a list of copy notify stateids. A copy notify structure makes a copy of the parent stateid and a clientid and will use it to look up the parent stateid during the READ request (suggested by Trond Myklebust <trond.myklebust@hammerspace.com>). At nfs4_put_stid() time, we walk the list of the associated copy notify stateids and delete them. Laundromat thread will traverse globally stored copy notify stateid in idr and notice if any haven't been referenced in the lease period, if so, it'll remove them. Return single netaddr to advertise to the copy. Suggested-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Andy Adamson <andros@netapp.com>
2019-10-05 04:34:26 +08:00
status = nfserrno(-ENOMEM);
cps = nfs4_alloc_init_cpntf_state(nn, stid);
if (!cps)
goto out;
NFSD: Rename the fields in copy_stateid_t Code maintenance: The name of the copy_stateid_t::sc_count field collides with the sc_count field in struct nfs4_stid, making the latter difficult to grep for when auditing stateid reference counting. No behavior change expected. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-23 01:10:35 +08:00
memcpy(&cn->cpn_cnr_stateid, &cps->cp_stateid.cs_stid, sizeof(stateid_t));
NFSD add COPY_NOTIFY operation Introducing the COPY_NOTIFY operation. Create a new unique stateid that will keep track of the copy state and the upcoming READs that will use that stateid. Each associated parent stateid has a list of copy notify stateids. A copy notify structure makes a copy of the parent stateid and a clientid and will use it to look up the parent stateid during the READ request (suggested by Trond Myklebust <trond.myklebust@hammerspace.com>). At nfs4_put_stid() time, we walk the list of the associated copy notify stateids and delete them. Laundromat thread will traverse globally stored copy notify stateid in idr and notice if any haven't been referenced in the lease period, if so, it'll remove them. Return single netaddr to advertise to the copy. Suggested-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Andy Adamson <andros@netapp.com>
2019-10-05 04:34:26 +08:00
memcpy(&cps->cp_p_stateid, &stid->sc_stateid, sizeof(stateid_t));
memcpy(&cps->cp_p_clid, &clp->cl_clientid, sizeof(clientid_t));
/* For now, only return one server address in cpn_src, the
* address used by the client to connect to this server.
*/
NFSD: Shrink size of struct nfsd4_copy_notify struct nfsd4_copy_notify is part of struct nfsd4_op, which resides in an 8-element array. sizeof(struct nfsd4_op): Before: /* size: 2208, cachelines: 35, members: 5 */ After: /* size: 1696, cachelines: 27, members: 5 */ Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:16 +08:00
cn->cpn_src->nl4_type = NL4_NETADDR;
NFSD add COPY_NOTIFY operation Introducing the COPY_NOTIFY operation. Create a new unique stateid that will keep track of the copy state and the upcoming READs that will use that stateid. Each associated parent stateid has a list of copy notify stateids. A copy notify structure makes a copy of the parent stateid and a clientid and will use it to look up the parent stateid during the READ request (suggested by Trond Myklebust <trond.myklebust@hammerspace.com>). At nfs4_put_stid() time, we walk the list of the associated copy notify stateids and delete them. Laundromat thread will traverse globally stored copy notify stateid in idr and notice if any haven't been referenced in the lease period, if so, it'll remove them. Return single netaddr to advertise to the copy. Suggested-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Andy Adamson <andros@netapp.com>
2019-10-05 04:34:26 +08:00
status = nfsd4_set_netaddr((struct sockaddr *)&rqstp->rq_daddr,
NFSD: Shrink size of struct nfsd4_copy_notify struct nfsd4_copy_notify is part of struct nfsd4_op, which resides in an 8-element array. sizeof(struct nfsd4_op): Before: /* size: 2208, cachelines: 35, members: 5 */ After: /* size: 1696, cachelines: 27, members: 5 */ Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:16 +08:00
&cn->cpn_src->u.nl4_addr);
NFSD add COPY_NOTIFY operation Introducing the COPY_NOTIFY operation. Create a new unique stateid that will keep track of the copy state and the upcoming READs that will use that stateid. Each associated parent stateid has a list of copy notify stateids. A copy notify structure makes a copy of the parent stateid and a clientid and will use it to look up the parent stateid during the READ request (suggested by Trond Myklebust <trond.myklebust@hammerspace.com>). At nfs4_put_stid() time, we walk the list of the associated copy notify stateids and delete them. Laundromat thread will traverse globally stored copy notify stateid in idr and notice if any haven't been referenced in the lease period, if so, it'll remove them. Return single netaddr to advertise to the copy. Suggested-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Andy Adamson <andros@netapp.com>
2019-10-05 04:34:26 +08:00
WARN_ON_ONCE(status);
if (status) {
nfs4_put_cpntf_state(nn, cps);
goto out;
}
out:
nfs4_put_stid(stid);
return status;
NFSD COPY_NOTIFY xdr Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-08-08 23:14:59 +08:00
}
nfsd: Add ALLOCATE support The ALLOCATE operation is used to preallocate space in a file. I can do this by using vfs_fallocate() to do the actual preallocation. ALLOCATE only returns a status indicator, so we don't need to write a special encode() function. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-11-08 03:44:26 +08:00
static __be32
nfsd4_fallocate(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
struct nfsd4_fallocate *fallocate, int flags)
{
nfsd4: skip unused assignment Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-11-07 06:44:03 +08:00
__be32 status;
nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp. Since we now presume that the struct file will be persistent in most cases, we can stop fiddling with the raparms in the read code. This also means that we don't really care about the rd_tmp_file field anymore. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-08-19 02:18:54 +08:00
struct nfsd_file *nf;
nfsd: Add ALLOCATE support The ALLOCATE operation is used to preallocate space in a file. I can do this by using vfs_fallocate() to do the actual preallocation. ALLOCATE only returns a status indicator, so we don't need to write a special encode() function. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-11-08 03:44:26 +08:00
nfsd: Pass filehandle to nfs4_preprocess_stateid_op() This will be needed so COPY can look up the saved_fh in addition to the current_fh. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 19:59:51 +08:00
status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
nfsd: Add ALLOCATE support The ALLOCATE operation is used to preallocate space in a file. I can do this by using vfs_fallocate() to do the actual preallocation. ALLOCATE only returns a status indicator, so we don't need to write a special encode() function. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-11-08 03:44:26 +08:00
&fallocate->falloc_stateid,
NFSD add COPY_NOTIFY operation Introducing the COPY_NOTIFY operation. Create a new unique stateid that will keep track of the copy state and the upcoming READs that will use that stateid. Each associated parent stateid has a list of copy notify stateids. A copy notify structure makes a copy of the parent stateid and a clientid and will use it to look up the parent stateid during the READ request (suggested by Trond Myklebust <trond.myklebust@hammerspace.com>). At nfs4_put_stid() time, we walk the list of the associated copy notify stateids and delete them. Laundromat thread will traverse globally stored copy notify stateid in idr and notice if any haven't been referenced in the lease period, if so, it'll remove them. Return single netaddr to advertise to the copy. Suggested-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Andy Adamson <andros@netapp.com>
2019-10-05 04:34:26 +08:00
WR_STATE, &nf, NULL);
NFSD: Clean up nfs4_preprocess_stateid_op() call sites Remove the lame-duck dprintk()s around nfs4_preprocess_stateid_op() call sites. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Tested-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: NeilBrown <neilb@suse.de>
2022-10-28 22:46:57 +08:00
if (status != nfs_ok)
nfsd: Add ALLOCATE support The ALLOCATE operation is used to preallocate space in a file. I can do this by using vfs_fallocate() to do the actual preallocation. ALLOCATE only returns a status indicator, so we don't need to write a special encode() function. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-11-08 03:44:26 +08:00
return status;
nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp. Since we now presume that the struct file will be persistent in most cases, we can stop fiddling with the raparms in the read code. This also means that we don't really care about the rd_tmp_file field anymore. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-08-19 02:18:54 +08:00
status = nfsd4_vfs_fallocate(rqstp, &cstate->current_fh, nf->nf_file,
nfsd: Add ALLOCATE support The ALLOCATE operation is used to preallocate space in a file. I can do this by using vfs_fallocate() to do the actual preallocation. ALLOCATE only returns a status indicator, so we don't need to write a special encode() function. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-11-08 03:44:26 +08:00
fallocate->falloc_offset,
fallocate->falloc_length,
flags);
nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp. Since we now presume that the struct file will be persistent in most cases, we can stop fiddling with the raparms in the read code. This also means that we don't really care about the rd_tmp_file field anymore. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-08-19 02:18:54 +08:00
nfsd_file_put(nf);
nfsd: Add ALLOCATE support The ALLOCATE operation is used to preallocate space in a file. I can do this by using vfs_fallocate() to do the actual preallocation. ALLOCATE only returns a status indicator, so we don't need to write a special encode() function. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-11-08 03:44:26 +08:00
return status;
}
nfsd: don't take nfsd4_copy ref for OP_OFFLOAD_STATUS We're not doing any blocking operations for OP_OFFLOAD_STATUS, so taking and putting a reference is a waste of effort. Take the client lock, search for the copy and fetch the wr_bytes_written field and return. Also, make find_async_copy a static function. Signed-off-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-01-19 01:31:34 +08:00
NFSD OFFLOAD_STATUS xdr Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:18 +08:00
static __be32
nfsd4_offload_status(struct svc_rqst *rqstp,
struct nfsd4_compound_state *cstate,
union nfsd4_op_u *u)
{
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
struct nfsd4_offload_status *os = &u->offload_status;
nfsd: don't take nfsd4_copy ref for OP_OFFLOAD_STATUS We're not doing any blocking operations for OP_OFFLOAD_STATUS, so taking and putting a reference is a waste of effort. Take the client lock, search for the copy and fetch the wr_bytes_written field and return. Also, make find_async_copy a static function. Signed-off-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-01-19 01:31:34 +08:00
__be32 status = nfs_ok;
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
struct nfsd4_copy *copy;
struct nfs4_client *clp = cstate->clp;
NFSD: Add COPY status code to OFFLOAD_STATUS response Clients that send an OFFLOAD_STATUS might want to distinguish between an async COPY operation that is still running, has completed successfully, or that has failed. The intention of this patch is to make NFSD behave like this: * Copy still running: OFFLOAD_STATUS returns NFS4_OK, the number of bytes copied so far, and an empty osr_status array * Copy completed successfully: OFFLOAD_STATUS returns NFS4_OK, the number of bytes copied, and an osr_status of NFS4_OK * Copy failed: OFFLOAD_STATUS returns NFS4_OK, the number of bytes copied, and an osr_status other than NFS4_OK * Copy operation lost, canceled, or otherwise unrecognized: OFFLOAD_STATUS returns NFS4ERR_BAD_STATEID NB: Though RFC 7862 Section 11.2 lists a small set of NFS status codes that are valid for OFFLOAD_STATUS, there do not seem to be any explicit spec limits on the status codes that may be returned in the osr_status field. At this time we have no unit tests for COPY and its brethren, as pynfs does not yet implement support for NFSv4.2. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-05-01 04:05:11 +08:00
os->completed = false;
nfsd: don't take nfsd4_copy ref for OP_OFFLOAD_STATUS We're not doing any blocking operations for OP_OFFLOAD_STATUS, so taking and putting a reference is a waste of effort. Take the client lock, search for the copy and fetch the wr_bytes_written field and return. Also, make find_async_copy a static function. Signed-off-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-01-19 01:31:34 +08:00
spin_lock(&clp->async_lock);
copy = find_async_copy_locked(clp, &os->stateid);
NFSD: Add COPY status code to OFFLOAD_STATUS response Clients that send an OFFLOAD_STATUS might want to distinguish between an async COPY operation that is still running, has completed successfully, or that has failed. The intention of this patch is to make NFSD behave like this: * Copy still running: OFFLOAD_STATUS returns NFS4_OK, the number of bytes copied so far, and an empty osr_status array * Copy completed successfully: OFFLOAD_STATUS returns NFS4_OK, the number of bytes copied, and an osr_status of NFS4_OK * Copy failed: OFFLOAD_STATUS returns NFS4_OK, the number of bytes copied, and an osr_status other than NFS4_OK * Copy operation lost, canceled, or otherwise unrecognized: OFFLOAD_STATUS returns NFS4ERR_BAD_STATEID NB: Though RFC 7862 Section 11.2 lists a small set of NFS status codes that are valid for OFFLOAD_STATUS, there do not seem to be any explicit spec limits on the status codes that may be returned in the osr_status field. At this time we have no unit tests for COPY and its brethren, as pynfs does not yet implement support for NFSv4.2. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-05-01 04:05:11 +08:00
if (copy) {
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
os->count = copy->cp_res.wr_bytes_written;
NFSD: Add COPY status code to OFFLOAD_STATUS response Clients that send an OFFLOAD_STATUS might want to distinguish between an async COPY operation that is still running, has completed successfully, or that has failed. The intention of this patch is to make NFSD behave like this: * Copy still running: OFFLOAD_STATUS returns NFS4_OK, the number of bytes copied so far, and an empty osr_status array * Copy completed successfully: OFFLOAD_STATUS returns NFS4_OK, the number of bytes copied, and an osr_status of NFS4_OK * Copy failed: OFFLOAD_STATUS returns NFS4_OK, the number of bytes copied, and an osr_status other than NFS4_OK * Copy operation lost, canceled, or otherwise unrecognized: OFFLOAD_STATUS returns NFS4ERR_BAD_STATEID NB: Though RFC 7862 Section 11.2 lists a small set of NFS status codes that are valid for OFFLOAD_STATUS, there do not seem to be any explicit spec limits on the status codes that may be returned in the osr_status field. At this time we have no unit tests for COPY and its brethren, as pynfs does not yet implement support for NFSv4.2. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-05-01 04:05:11 +08:00
if (test_bit(NFSD4_COPY_F_COMPLETED, &copy->cp_flags)) {
os->completed = true;
os->status = copy->nfserr;
}
} else
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
status = nfserr_bad_stateid;
nfsd: don't take nfsd4_copy ref for OP_OFFLOAD_STATUS We're not doing any blocking operations for OP_OFFLOAD_STATUS, so taking and putting a reference is a waste of effort. Take the client lock, search for the copy and fetch the wr_bytes_written field and return. Also, make find_async_copy a static function. Signed-off-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-01-19 01:31:34 +08:00
spin_unlock(&clp->async_lock);
NFSD introduce async copy feature Upon receiving a request for async copy, create a new kthread. If we get asynchronous request, make sure to copy the needed arguments/state from the stack before starting the copy. Then start the thread and reply back to the client indicating copy is asynchronous. nfsd_copy_file_range() will copy in a loop over the total number of bytes is needed to copy. In case a failure happens in the middle, we ignore the error and return how much we copied so far. Once done creating a workitem for the callback workqueue and send CB_OFFLOAD with the results. The lifetime of the copy stateid is bound to the vfs copy. This way we don't need to keep the nfsd_net structure for the callback. We could keep it around longer so that an OFFLOAD_STATUS that came late would still get results, but clients should be able to deal without that. We handle OFFLOAD_CANCEL by sending a signal to the copy thread and calling kthread_stop. A client should cancel any ongoing copies before calling DESTROY_CLIENT; if not, we return a CLIENT_BUSY error. If the client is destroyed for some other reason (lease expiration, or server shutdown), we must clean up any ongoing copies ourselves. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> [colin.king@canonical.com: fix leak in error case] [bfields@fieldses.org: remove signalling, merge patches] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:20 +08:00
return status;
NFSD OFFLOAD_STATUS xdr Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:18 +08:00
}
nfsd: Add ALLOCATE support The ALLOCATE operation is used to preallocate space in a file. I can do this by using vfs_fallocate() to do the actual preallocation. ALLOCATE only returns a status indicator, so we don't need to write a special encode() function. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-11-08 03:44:26 +08:00
static __be32
nfsd4_allocate(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
nfsd: Add ALLOCATE support The ALLOCATE operation is used to preallocate space in a file. I can do this by using vfs_fallocate() to do the actual preallocation. ALLOCATE only returns a status indicator, so we don't need to write a special encode() function. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-11-08 03:44:26 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
return nfsd4_fallocate(rqstp, cstate, &u->allocate, 0);
nfsd: Add ALLOCATE support The ALLOCATE operation is used to preallocate space in a file. I can do this by using vfs_fallocate() to do the actual preallocation. ALLOCATE only returns a status indicator, so we don't need to write a special encode() function. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-11-08 03:44:26 +08:00
}
nfsd: Add DEALLOCATE support DEALLOCATE only returns a status value, meaning we can use the noop() xdr encoder to reply to the client. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-11-08 03:44:27 +08:00
static __be32
nfsd4_deallocate(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
nfsd: Add DEALLOCATE support DEALLOCATE only returns a status value, meaning we can use the noop() xdr encoder to reply to the client. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-11-08 03:44:27 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
return nfsd4_fallocate(rqstp, cstate, &u->deallocate,
nfsd: Add DEALLOCATE support DEALLOCATE only returns a status value, meaning we can use the noop() xdr encoder to reply to the client. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-11-08 03:44:27 +08:00
FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE);
}
NFSD: Implement SEEK This patch adds server support for the NFS v4.2 operation SEEK, which returns the position of the next hole or data segment in a file. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-09-27 01:58:27 +08:00
static __be32
nfsd4_seek(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
NFSD: Implement SEEK This patch adds server support for the NFS v4.2 operation SEEK, which returns the position of the next hole or data segment in a file. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-09-27 01:58:27 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_seek *seek = &u->seek;
NFSD: Implement SEEK This patch adds server support for the NFS v4.2 operation SEEK, which returns the position of the next hole or data segment in a file. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-09-27 01:58:27 +08:00
int whence;
__be32 status;
nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp. Since we now presume that the struct file will be persistent in most cases, we can stop fiddling with the raparms in the read code. This also means that we don't really care about the rd_tmp_file field anymore. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-08-19 02:18:54 +08:00
struct nfsd_file *nf;
NFSD: Implement SEEK This patch adds server support for the NFS v4.2 operation SEEK, which returns the position of the next hole or data segment in a file. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-09-27 01:58:27 +08:00
nfsd: Pass filehandle to nfs4_preprocess_stateid_op() This will be needed so COPY can look up the saved_fh in addition to the current_fh. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 19:59:51 +08:00
status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
NFSD: Implement SEEK This patch adds server support for the NFS v4.2 operation SEEK, which returns the position of the next hole or data segment in a file. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-09-27 01:58:27 +08:00
&seek->seek_stateid,
NFSD add COPY_NOTIFY operation Introducing the COPY_NOTIFY operation. Create a new unique stateid that will keep track of the copy state and the upcoming READs that will use that stateid. Each associated parent stateid has a list of copy notify stateids. A copy notify structure makes a copy of the parent stateid and a clientid and will use it to look up the parent stateid during the READ request (suggested by Trond Myklebust <trond.myklebust@hammerspace.com>). At nfs4_put_stid() time, we walk the list of the associated copy notify stateids and delete them. Laundromat thread will traverse globally stored copy notify stateid in idr and notice if any haven't been referenced in the lease period, if so, it'll remove them. Return single netaddr to advertise to the copy. Suggested-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Andy Adamson <andros@netapp.com>
2019-10-05 04:34:26 +08:00
RD_STATE, &nf, NULL);
NFSD: Clean up nfs4_preprocess_stateid_op() call sites Remove the lame-duck dprintk()s around nfs4_preprocess_stateid_op() call sites. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Tested-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: NeilBrown <neilb@suse.de>
2022-10-28 22:46:57 +08:00
if (status)
NFSD: Implement SEEK This patch adds server support for the NFS v4.2 operation SEEK, which returns the position of the next hole or data segment in a file. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-09-27 01:58:27 +08:00
return status;
switch (seek->seek_whence) {
case NFS4_CONTENT_DATA:
whence = SEEK_DATA;
break;
case NFS4_CONTENT_HOLE:
whence = SEEK_HOLE;
break;
default:
status = nfserr_union_notsupp;
goto out;
}
/*
* Note: This call does change file->f_pos, but nothing in NFSD
* should ever file->f_pos.
*/
nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp. Since we now presume that the struct file will be persistent in most cases, we can stop fiddling with the raparms in the read code. This also means that we don't really care about the rd_tmp_file field anymore. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-08-19 02:18:54 +08:00
seek->seek_pos = vfs_llseek(nf->nf_file, seek->seek_offset, whence);
NFSD: Implement SEEK This patch adds server support for the NFS v4.2 operation SEEK, which returns the position of the next hole or data segment in a file. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-09-27 01:58:27 +08:00
if (seek->seek_pos < 0)
status = nfserrno(seek->seek_pos);
nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp. Since we now presume that the struct file will be persistent in most cases, we can stop fiddling with the raparms in the read code. This also means that we don't really care about the rd_tmp_file field anymore. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-08-19 02:18:54 +08:00
else if (seek->seek_pos >= i_size_read(file_inode(nf->nf_file)))
NFSD: Implement SEEK This patch adds server support for the NFS v4.2 operation SEEK, which returns the position of the next hole or data segment in a file. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-09-27 01:58:27 +08:00
seek->seek_eof = true;
out:
nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp. Since we now presume that the struct file will be persistent in most cases, we can stop fiddling with the raparms in the read code. This also means that we don't really care about the rd_tmp_file field anymore. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-08-19 02:18:54 +08:00
nfsd_file_put(nf);
NFSD: Implement SEEK This patch adds server support for the NFS v4.2 operation SEEK, which returns the position of the next hole or data segment in a file. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-09-27 01:58:27 +08:00
return status;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* This routine never returns NFS_OK! If there are no other errors, it
* will return NFSERR_SAME or NFSERR_NOT_SAME depending on whether the
* attributes matched. VERIFY is implemented by mapping NFSERR_SAME
* to NFS_OK after the call; NVERIFY by mapping NFSERR_NOT_SAME to NFS_OK.
*/
[PATCH] nfsd: nfs4 code returns error values in net-endian Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:28:59 +08:00
static __be32
[PATCH] knfsd: nfsd4: make verify and nverify wrappers Make wrappers for verify and nverify, for consistency with other ops. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:31 +08:00
_nfsd4_verify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
struct nfsd4_verify *verify)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
[PATCH] xdr annotations: NFSv4 server Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:28:58 +08:00
__be32 *buf, *p;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
int count;
[PATCH] nfsd: nfs4 code returns error values in net-endian Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:28:59 +08:00
__be32 status;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
nfsd: rename MAY_ flags Rename nfsd_permission() specific MAY_* flags to NFSD_MAY_* to make it clear, that these are not used outside nfsd, and to avoid name and number space conflicts with the VFS. [comment from hch: rename MAY_READ, MAY_WRITE and MAY_EXEC as well] Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-06-16 19:20:29 +08:00
status = fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_NOP);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (status)
return status;
NFSv4: do exact check about attribute specified Server should return NFS4ERR_ATTRNOTSUPP if an attribute specified is not supported in current environment. Operations CREATE, NVERIFY, OPEN, SETATTR and VERIFY should do this check. This bug is found when do newpynfs tests. The names of the tests that failed are following: CR12 NVF7a NVF7b NVF7c NVF7d NVF7f NVF7r NVF7s OPEN15 VF7a VF7b VF7c VF7d VF7f VF7r VF7s Add function do_check_fattr() to do exact check: 1, Check attribute specified is supported by the NFSv4 server or not. 2, Check FATTR4_WORD0_ACL & FATTR4_WORD0_FS_LOCATIONS are supported in current environment or not. 3, Check attribute specified is writable or not. step 1 and 3 are done in function nfsd4_decode_fattr() but removed to this function now. Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-05-16 16:22:31 +08:00
status = check_attr_support(rqstp, cstate, verify->ve_bmval, NULL);
if (status)
return status;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if ((verify->ve_bmval[0] & FATTR4_WORD0_RDATTR_ERROR)
|| (verify->ve_bmval[1] & NFSD_WRITEONLY_ATTRS_WORD1))
return nfserr_inval;
if (verify->ve_attrlen & 3)
return nfserr_inval;
/* count in words:
* bitmap_len(1) + bitmap(2) + attr_len(1) = 4
*/
count = 4 + (verify->ve_attrlen >> 2);
buf = kmalloc(count << 2, GFP_KERNEL);
if (!buf)
nfsd4: stop using nfserr_resource for transitory errors The server is returning nfserr_resource for both permanent errors and for errors (like allocation failures) that might be resolved by retrying later. Save nfserr_resource for the former and use delay/jukebox for the latter. Cc: stable@kernel.org Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-11 07:07:33 +08:00
return nfserr_jukebox;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
nfsd4: simplify nfsd4_encode_fattr interface slightly It seems slightly simpler to make nfsd4_encode_fattr rather than its callers responsible for advancing the write pointer on success. (Also: the count == 0 check in the verify case looks superfluous. Running out of buffer space is really the only reason fattr encoding should fail with eresource.) Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-12-15 06:57:50 +08:00
p = buf;
nfsd4: tweak nfsd4_encode_getattr to take xdr_stream Just change the nfsd4_encode_getattr api. Not changing any code or adding any new functionality yet. Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-08-27 04:04:46 +08:00
status = nfsd4_encode_fattr_to_buf(&p, count, &cstate->current_fh,
[PATCH] knfsd: nfsd4: pass saved and current fh together into nfsd4 operations Pass the saved and current filehandles together into all the nfsd4 compound operations. I want a unified interface to these operations so we can just call them by pointer and throw out the huge switch statement. Also I'll eventually want a structure like this--that holds the state used during compound processing--for deferral. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:27 +08:00
cstate->current_fh.fh_export,
nfsd4: tweak nfsd4_encode_getattr to take xdr_stream Just change the nfsd4_encode_getattr api. Not changing any code or adding any new functionality yet. Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-08-27 04:04:46 +08:00
cstate->current_fh.fh_dentry,
verify->ve_bmval,
nfsd: Allow AIX client to read dir containing mountpoints This patch addresses a compatibility issue with a Linux NFS server and AIX NFS client. I have exported /export as fsid=0 with sec=krb5:krb5i I have mount --bind /home onto /export/home I have exported /export/home with sec=krb5i The AIX client mounts / -o sec=krb5:krb5i onto /mnt If I do an ls /mnt, the AIX client gets a permission error. Looking at the network traceIwe see a READDIR looking for attributes FATTR4_RDATTR_ERROR and FATTR4_MOUNTED_ON_FILEID. The response gives a NFS4ERR_WRONGSEC which the AIX client is not expecting. Since the AIX client is only asking for an attribute that is an attribute of the parent file system (pseudo root in my example), it seems reasonable that there should not be an error. In discussing this issue with Bruce Fields, I initially proposed ignoring the error in nfsd4_encode_dirent_fattr() if all that was being asked for was FATTR4_RDATTR_ERROR and FATTR4_MOUNTED_ON_FILEID, however, Bruce suggested that we avoid calling cross_mnt() if only these attributes are requested. The following patch implements bypassing cross_mnt() if only FATTR4_RDATTR_ERROR and FATTR4_MOUNTED_ON_FILEID are called. Since there is some complexity in the code in nfsd4_encode_fattr(), I didn't want to duplicate code (and introduce a maintenance nightmare), so I added a parameter to nfsd4_encode_fattr() that indicates whether it should ignore cross mounts and simply fill in the attribute using the passed in dentry as opposed to it's parent. Signed-off-by: Frank Filz <ffilzlnx@us.ibm.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2007-11-28 03:34:05 +08:00
rqstp, 0);
nfsd4: better VERIFY comment This confuses me every time. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-08-22 03:32:50 +08:00
/*
* If nfsd4_encode_fattr() ran out of space, assume that's because
* the attributes are longer (hence different) than those given:
*/
nfsd4: simplify nfsd4_encode_fattr interface slightly It seems slightly simpler to make nfsd4_encode_fattr rather than its callers responsible for advancing the write pointer on success. (Also: the count == 0 check in the verify case looks superfluous. Running out of buffer space is really the only reason fattr encoding should fail with eresource.) Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-12-15 06:57:50 +08:00
if (status == nfserr_resource)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
status = nfserr_not_same;
if (status)
goto out_kfree;
nfsd: dynamically skip encoded fattr bitmap in _nfsd4_verify _nfsd4_verify currently skips 3 words from the encoded buffer begining. With support for 3-word attr bitmaps in nfsd41, nfsd4_encode_fattr may encode 1, 2, or 3 words, and not always 2 as it used to be, hence we need to find out where to skip using the encoded bitmap length. Note: This patch may be applied over pre-nfsd41 nfsd. Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:29:08 +08:00
/* skip bitmap */
p = buf + 1 + ntohl(buf[0]);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
status = nfserr_not_same;
if (ntohl(*p++) != verify->ve_attrlen)
goto out_kfree;
if (!memcmp(p, verify->ve_attrval, verify->ve_attrlen))
status = nfserr_same;
out_kfree:
kfree(buf);
return status;
}
[PATCH] knfsd: nfsd4: make verify and nverify wrappers Make wrappers for verify and nverify, for consistency with other ops. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:31 +08:00
static __be32
nfsd4_nverify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
[PATCH] knfsd: nfsd4: make verify and nverify wrappers Make wrappers for verify and nverify, for consistency with other ops. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:31 +08:00
{
__be32 status;
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
status = _nfsd4_verify(rqstp, cstate, &u->verify);
[PATCH] knfsd: nfsd4: make verify and nverify wrappers Make wrappers for verify and nverify, for consistency with other ops. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:31 +08:00
return status == nfserr_not_same ? nfs_ok : status;
}
static __be32
nfsd4_verify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
union nfsd4_op_u *u)
[PATCH] knfsd: nfsd4: make verify and nverify wrappers Make wrappers for verify and nverify, for consistency with other ops. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:31 +08:00
{
__be32 status;
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
status = _nfsd4_verify(rqstp, cstate, &u->nverify);
[PATCH] knfsd: nfsd4: make verify and nverify wrappers Make wrappers for verify and nverify, for consistency with other ops. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:31 +08:00
return status == nfserr_same ? nfs_ok : status;
}
nfsd: trivial GET_DIR_DELEGATION support This adds basic infrastructure for handing GET_DIR_DELEGATION calls from clients, including the decoders and encoders. For now, it always just returns NFS4_OK + GDD4_UNAVAIL. Eventually clients may start sending this operation, and it's better if we can return GDD4_UNAVAIL instead of having to abort the whole compound. Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-03-23 21:25:54 +08:00
static __be32
nfsd4_get_dir_delegation(struct svc_rqst *rqstp,
struct nfsd4_compound_state *cstate,
union nfsd4_op_u *u)
{
struct nfsd4_get_dir_delegation *gdd = &u->get_dir_delegation;
/*
* RFC 8881, section 18.39.3 says:
*
* "The server may refuse to grant the delegation. In that case, the
* server will return NFS4ERR_DIRDELEG_UNAVAIL."
*
* This is sub-optimal, since it means that the server would need to
* abort compound processing just because the delegation wasn't
* available. RFC8881bis should change this to allow the server to
* return NFS4_OK with a non-fatal status of GDD4_UNAVAIL in this
* situation.
*/
gdd->gddrnf_status = GDD4_UNAVAIL;
return nfs_ok;
}
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
#ifdef CONFIG_NFSD_PNFS
static const struct nfsd4_layout_ops *
nfsd4_layout_verify(struct svc_export *exp, unsigned int layout_type)
{
nfsd: allow nfsd to advertise multiple layout types If the underlying filesystem supports multiple layout types, then there is little reason not to advertise that fact to clients and let them choose what type to use. Turn the ex_layout_type field into a bitfield. For each supported layout type, we set a bit in that field. When the client requests a layout, ensure that the bit for that layout type is set. When the client requests attributes, send back a list of supported types. Signed-off-by: Jeff Layton <jlayton@poochiereds.net> Reviewed-by: Weston Andros Adamson <dros@primarydata.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-07-11 03:55:58 +08:00
if (!exp->ex_layout_types) {
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
dprintk("%s: export does not support pNFS\n", __func__);
return NULL;
}
nfsd: fix undefined behavior in nfsd4_layout_verify UBSAN: Undefined behaviour in fs/nfsd/nfs4proc.c:1262:34 shift exponent 128 is too large for 32-bit type 'int' Depending on compiler+architecture, this may cause the check for layout_type to succeed for overly large values (which seems to be the case with amd64). The large value will be later used in de-referencing nfsd4_layout_ops for function pointers. Reported-by: Jani Tuovila <tuovila@synopsys.com> Signed-off-by: Ari Kauppi <ari@synopsys.com> [colin.king@canonical.com: use LAYOUT_TYPE_MAX instead of 32] Cc: stable@vger.kernel.org Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-05-06 04:07:55 +08:00
if (layout_type >= LAYOUT_TYPE_MAX ||
!(exp->ex_layout_types & (1 << layout_type))) {
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
dprintk("%s: layout type %d not supported\n",
__func__, layout_type);
return NULL;
}
return nfsd4_layout_ops[layout_type];
}
static __be32
nfsd4_getdeviceinfo(struct svc_rqst *rqstp,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_compound_state *cstate, union nfsd4_op_u *u)
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_getdeviceinfo *gdp = &u->getdeviceinfo;
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
const struct nfsd4_layout_ops *ops;
struct nfsd4_deviceid_map *map;
struct svc_export *exp;
__be32 nfserr;
dprintk("%s: layout_type %u dev_id [0x%llx:0x%x] maxcnt %u\n",
__func__,
gdp->gd_layout_type,
gdp->gd_devid.fsid_idx, gdp->gd_devid.generation,
gdp->gd_maxcount);
map = nfsd4_find_devid_map(gdp->gd_devid.fsid_idx);
if (!map) {
dprintk("%s: couldn't find device ID to export mapping!\n",
__func__);
return nfserr_noent;
}
nfsd: Don't pass all of rqst into rqst_exp_find() Rather than passing the whole rqst, pass the pieces that are actually needed. This makes the inputs to rqst_exp_find() more obvious. Signed-off-by: NeilBrown <neilb@suse.de> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-07-26 10:21:30 +08:00
exp = rqst_exp_find(&rqstp->rq_chandle, SVC_NET(rqstp),
rqstp->rq_client, rqstp->rq_gssclient,
map->fsid_type, map->fsid);
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
if (IS_ERR(exp)) {
dprintk("%s: could not find device id\n", __func__);
return nfserr_noent;
}
nfserr = nfserr_layoutunavailable;
ops = nfsd4_layout_verify(exp, gdp->gd_layout_type);
if (!ops)
goto out;
nfserr = nfs_ok;
nfsd: add SCSI layout support This is a simple extension to the block layout driver to use SCSI persistent reservations for access control and fencing, as well as SCSI VPD pages for device identification. For this we need to pass the nfs4_client to the proc_getdeviceinfo method to generate the reservation key, and add a new fence_client method to allow for fence actions in the layout driver. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-03-05 03:46:17 +08:00
if (gdp->gd_maxcount != 0) {
nfserr = ops->proc_getdeviceinfo(exp->ex_path.mnt->mnt_sb,
nfsd: cstate->session->se_client -> cstate->clp I'm not sure why we're writing this out the hard way in so many places. Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2021-01-22 06:57:45 +08:00
rqstp, cstate->clp, gdp);
nfsd: add SCSI layout support This is a simple extension to the block layout driver to use SCSI persistent reservations for access control and fencing, as well as SCSI VPD pages for device identification. For this we need to pass the nfs4_client to the proc_getdeviceinfo method to generate the reservation key, and add a new fence_client method to allow for fence actions in the layout driver. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-03-05 03:46:17 +08:00
}
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
gdp->gd_notify_types &= ops->notify_types;
out:
NFSD: Put exports after nfsd4_layout_verify fail Fix commit 9cf514ccfa (nfsd: implement pNFS operations). Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2015-03-15 23:12:15 +08:00
exp_put(exp);
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
return nfserr;
}
nfsd4: define ->op_release for compound ops Run a separate ->op_release function if necessary instead of depending on the xdr encoder to do this. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-05-06 05:09:37 +08:00
static void
nfsd4_getdeviceinfo_release(union nfsd4_op_u *u)
{
kfree(u->getdeviceinfo.gd_device);
}
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
static __be32
nfsd4_layoutget(struct svc_rqst *rqstp,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_compound_state *cstate, union nfsd4_op_u *u)
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_layoutget *lgp = &u->layoutget;
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
struct svc_fh *current_fh = &cstate->current_fh;
const struct nfsd4_layout_ops *ops;
struct nfs4_layout_stateid *ls;
__be32 nfserr;
NFSD: Support write delegations in LAYOUTGET I noticed LAYOUTGET(LAYOUTIOMODE4_RW) returning NFS4ERR_ACCESS unexpectedly. The NFS client had created a file with mode 0444, and the server had returned a write delegation on the OPEN(CREATE). The client was requesting a RW layout using the write delegation stateid so that it could flush file modifications. Creating a read-only file does not seem to be problematic for NFSv4.1 without pNFS, so I began looking at NFSD's implementation of LAYOUTGET. The failure was because fh_verify() was doing a permission check as part of verifying the FH presented during the LAYOUTGET. It uses the loga_iomode value to specify the @accmode argument to fh_verify(). fh_verify(MAY_WRITE) on a file whose mode is 0444 fails with -EACCES. To permit LAYOUT* operations in this case, add OWNER_OVERRIDE when checking the access permission of the incoming file handle for LAYOUTGET and LAYOUTCOMMIT. Cc: Christoph Hellwig <hch@lst.de> Cc: stable@vger.kernel.org # v6.6+ Message-Id: 4E9C0D74-A06D-4DC3-A48A-73034DC40395@oracle.com Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-06-12 03:36:46 +08:00
int accmode = NFSD_MAY_READ_IF_EXEC | NFSD_MAY_OWNER_OVERRIDE;
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
switch (lgp->lg_seg.iomode) {
case IOMODE_READ:
nfsd4: permit layoutget of executable-only files Clients must be able to read a file in order to execute it, and for pNFS that means the client needs to be able to perform a LAYOUTGET on the file. This behavior for executable-only files was added for OPEN in commit a043226bc140 "nfsd4: permit read opens of executable-only files". This fixes up xfstests generic/126 on block/scsi layouts. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-12-19 22:35:25 +08:00
accmode |= NFSD_MAY_READ;
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
break;
case IOMODE_RW:
nfsd4: permit layoutget of executable-only files Clients must be able to read a file in order to execute it, and for pNFS that means the client needs to be able to perform a LAYOUTGET on the file. This behavior for executable-only files was added for OPEN in commit a043226bc140 "nfsd4: permit read opens of executable-only files". This fixes up xfstests generic/126 on block/scsi layouts. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-12-19 22:35:25 +08:00
accmode |= NFSD_MAY_READ | NFSD_MAY_WRITE;
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
break;
default:
dprintk("%s: invalid iomode %d\n",
__func__, lgp->lg_seg.iomode);
nfserr = nfserr_badiomode;
goto out;
}
nfserr = fh_verify(rqstp, current_fh, 0, accmode);
if (nfserr)
goto out;
nfserr = nfserr_layoutunavailable;
ops = nfsd4_layout_verify(current_fh->fh_export, lgp->lg_layout_type);
if (!ops)
goto out;
/*
* Verify minlength and range as per RFC5661:
* o If loga_length is less than loga_minlength,
* the metadata server MUST return NFS4ERR_INVAL.
* o If the sum of loga_offset and loga_minlength exceeds
* NFS4_UINT64_MAX, and loga_minlength is not
* NFS4_UINT64_MAX, the error NFS4ERR_INVAL MUST result.
* o If the sum of loga_offset and loga_length exceeds
* NFS4_UINT64_MAX, and loga_length is not NFS4_UINT64_MAX,
* the error NFS4ERR_INVAL MUST result.
*/
nfserr = nfserr_inval;
if (lgp->lg_seg.length < lgp->lg_minlength ||
(lgp->lg_minlength != NFS4_MAX_UINT64 &&
lgp->lg_minlength > NFS4_MAX_UINT64 - lgp->lg_seg.offset) ||
(lgp->lg_seg.length != NFS4_MAX_UINT64 &&
lgp->lg_seg.length > NFS4_MAX_UINT64 - lgp->lg_seg.offset))
goto out;
if (lgp->lg_seg.length == 0)
goto out;
nfserr = nfsd4_preprocess_layout_stateid(rqstp, cstate, &lgp->lg_sid,
true, lgp->lg_layout_type, &ls);
nfsd: add trace events For now just a few simple events to trace the layout stateid lifetime, but these already were enough to find several bugs in the Linux client layout stateid handling. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-08-17 08:02:22 +08:00
if (nfserr) {
nfsd: Add "nfsd_" to trace point names Follow naming convention used in client and in sunrpc layers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-03-27 22:53:11 +08:00
trace_nfsd_layout_get_lookup_fail(&lgp->lg_sid);
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
goto out;
nfsd: add trace events For now just a few simple events to trace the layout stateid lifetime, but these already were enough to find several bugs in the Linux client layout stateid handling. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-08-17 08:02:22 +08:00
}
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
nfsd: implement pNFS layout recalls Add support to issue layout recalls to clients. For now we only support full-file recalls to get a simple and stable implementation. This allows to embedd a nfsd4_callback structure in the layout_state and thus avoid any memory allocations under spinlocks during a recall. For normal use cases that do not intent to share a single file between multiple clients this implementation is fully sufficient. To ensure layouts are recalled on local filesystem access each layout state registers a new FL_LAYOUT lease with the kernel file locking code, which filesystems that support pNFS exports that require recalls need to break on conflicting access patterns. The XDR code is based on the old pNFS server implementation by Andy Adamson, Benny Halevy, Boaz Harrosh, Dean Hildebrand, Fred Isaman, Marc Eshel, Mike Sager and Ricardo Labiaga. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-09-23 18:38:48 +08:00
nfserr = nfserr_recallconflict;
if (atomic_read(&ls->ls_stid.sc_file->fi_lo_recalls))
goto out_put_stid;
VFS: normal filesystems (and lustre): d_inode() annotations that's the bulk of filesystem drivers dealing with inodes of their own Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-03-18 06:25:59 +08:00
nfserr = ops->proc_layoutget(d_inode(current_fh->fh_dentry),
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
current_fh, lgp);
if (nfserr)
goto out_put_stid;
nfserr = nfsd4_insert_layout(lgp, ls);
out_put_stid:
nfsd: serialize layout stateid morphing operations In order to allow the client to make a sane determination of what happened with racing LAYOUTGET/LAYOUTRETURN/CB_LAYOUTRECALL calls, we must ensure that the seqids return accurately represent the order of operations. The simplest way to do that is to ensure that operations on a single stateid are serialized. This patch adds a mutex to the layout stateid, and locks it when checking the layout stateid's seqid. The mutex is held over the entire operation and released after the seqid is bumped. Note that in the case of CB_LAYOUTRECALL we must move the increment of the seqid and setting into a new cb "prepare" operation. The lease infrastructure will call the lm_break callback with a spinlock held, so and we can't take the mutex in that codepath. Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2015-09-17 19:58:24 +08:00
mutex_unlock(&ls->ls_mutex);
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
nfs4_put_stid(&ls->ls_stid);
out:
return nfserr;
}
nfsd4: define ->op_release for compound ops Run a separate ->op_release function if necessary instead of depending on the xdr encoder to do this. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-05-06 05:09:37 +08:00
static void
nfsd4_layoutget_release(union nfsd4_op_u *u)
{
kfree(u->layoutget.lg_content);
}
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
static __be32
nfsd4_layoutcommit(struct svc_rqst *rqstp,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_compound_state *cstate, union nfsd4_op_u *u)
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_layoutcommit *lcp = &u->layoutcommit;
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
const struct nfsd4_layout_seg *seg = &lcp->lc_seg;
struct svc_fh *current_fh = &cstate->current_fh;
const struct nfsd4_layout_ops *ops;
loff_t new_size = lcp->lc_last_wr + 1;
struct inode *inode;
struct nfs4_layout_stateid *ls;
__be32 nfserr;
NFSD: Support write delegations in LAYOUTGET I noticed LAYOUTGET(LAYOUTIOMODE4_RW) returning NFS4ERR_ACCESS unexpectedly. The NFS client had created a file with mode 0444, and the server had returned a write delegation on the OPEN(CREATE). The client was requesting a RW layout using the write delegation stateid so that it could flush file modifications. Creating a read-only file does not seem to be problematic for NFSv4.1 without pNFS, so I began looking at NFSD's implementation of LAYOUTGET. The failure was because fh_verify() was doing a permission check as part of verifying the FH presented during the LAYOUTGET. It uses the loga_iomode value to specify the @accmode argument to fh_verify(). fh_verify(MAY_WRITE) on a file whose mode is 0444 fails with -EACCES. To permit LAYOUT* operations in this case, add OWNER_OVERRIDE when checking the access permission of the incoming file handle for LAYOUTGET and LAYOUTCOMMIT. Cc: Christoph Hellwig <hch@lst.de> Cc: stable@vger.kernel.org # v6.6+ Message-Id: 4E9C0D74-A06D-4DC3-A48A-73034DC40395@oracle.com Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-06-12 03:36:46 +08:00
nfserr = fh_verify(rqstp, current_fh, 0,
NFSD_MAY_WRITE | NFSD_MAY_OWNER_OVERRIDE);
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
if (nfserr)
goto out;
nfserr = nfserr_layoutunavailable;
ops = nfsd4_layout_verify(current_fh->fh_export, lcp->lc_layout_type);
if (!ops)
goto out;
VFS: normal filesystems (and lustre): d_inode() annotations that's the bulk of filesystem drivers dealing with inodes of their own Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-03-18 06:25:59 +08:00
inode = d_inode(current_fh->fh_dentry);
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
nfserr = nfserr_inval;
if (new_size <= seg->offset) {
dprintk("pnfsd: last write before layout segment\n");
goto out;
}
if (new_size > seg->offset + seg->length) {
dprintk("pnfsd: last write beyond layout segment\n");
goto out;
}
if (!lcp->lc_newoffset && new_size > i_size_read(inode)) {
dprintk("pnfsd: layoutcommit beyond EOF\n");
goto out;
}
nfserr = nfsd4_preprocess_layout_stateid(rqstp, cstate, &lcp->lc_sid,
false, lcp->lc_layout_type,
&ls);
if (nfserr) {
nfsd: Add "nfsd_" to trace point names Follow naming convention used in client and in sunrpc layers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-03-27 22:53:11 +08:00
trace_nfsd_layout_commit_lookup_fail(&lcp->lc_sid);
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
/* fixup error code as per RFC5661 */
if (nfserr == nfserr_bad_stateid)
nfserr = nfserr_badlayout;
goto out;
}
nfsd: serialize layout stateid morphing operations In order to allow the client to make a sane determination of what happened with racing LAYOUTGET/LAYOUTRETURN/CB_LAYOUTRECALL calls, we must ensure that the seqids return accurately represent the order of operations. The simplest way to do that is to ensure that operations on a single stateid are serialized. This patch adds a mutex to the layout stateid, and locks it when checking the layout stateid's seqid. The mutex is held over the entire operation and released after the seqid is bumped. Note that in the case of CB_LAYOUTRECALL we must move the increment of the seqid and setting into a new cb "prepare" operation. The lease infrastructure will call the lm_break callback with a spinlock held, so and we can't take the mutex in that codepath. Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2015-09-17 19:58:24 +08:00
/* LAYOUTCOMMIT does not require any serialization */
mutex_unlock(&ls->ls_mutex);
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
if (new_size > i_size_read(inode)) {
NFSD: Clean up nfsd4_encode_layoutcommit() Adopt the use of conventional XDR utility functions. Restructure the encoder to better align with the XDR definition of the result. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-09-25 21:28:04 +08:00
lcp->lc_size_chg = true;
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
lcp->lc_newsize = new_size;
} else {
NFSD: Clean up nfsd4_encode_layoutcommit() Adopt the use of conventional XDR utility functions. Restructure the encoder to better align with the XDR definition of the result. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-09-25 21:28:04 +08:00
lcp->lc_size_chg = false;
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
}
nfsd: Set lc_size_chg before ops->proc_layoutcommit After proc_layoutcommit success, i_size_read(inode) always >= new_size. Just set lc_size_chg before proc_layoutcommit, if proc_layoutcommit failed, nfsd will skip the lc_size_chg, so it's no harm. Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2015-07-07 10:12:03 +08:00
nfserr = ops->proc_layoutcommit(inode, lcp);
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
nfs4_put_stid(&ls->ls_stid);
out:
return nfserr;
}
static __be32
nfsd4_layoutreturn(struct svc_rqst *rqstp,
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_compound_state *cstate, union nfsd4_op_u *u)
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
{
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
struct nfsd4_layoutreturn *lrp = &u->layoutreturn;
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
struct svc_fh *current_fh = &cstate->current_fh;
__be32 nfserr;
nfserr = fh_verify(rqstp, current_fh, 0, NFSD_MAY_NOP);
if (nfserr)
goto out;
nfserr = nfserr_layoutunavailable;
if (!nfsd4_layout_verify(current_fh->fh_export, lrp->lr_layout_type))
goto out;
switch (lrp->lr_seg.iomode) {
case IOMODE_READ:
case IOMODE_RW:
case IOMODE_ANY:
break;
default:
dprintk("%s: invalid iomode %d\n", __func__,
lrp->lr_seg.iomode);
nfserr = nfserr_inval;
goto out;
}
switch (lrp->lr_return_type) {
case RETURN_FILE:
nfserr = nfsd4_return_file_layouts(rqstp, cstate, lrp);
break;
case RETURN_FSID:
case RETURN_ALL:
nfserr = nfsd4_return_client_layouts(rqstp, cstate, lrp);
break;
default:
dprintk("%s: invalid return_type %d\n", __func__,
lrp->lr_return_type);
nfserr = nfserr_inval;
break;
}
out:
return nfserr;
}
#endif /* CONFIG_NFSD_PNFS */
nfsd: implement the xattr functions and en/decode logic Implement the main entry points for the *XATTR operations. Add functions to calculate the reply size for the user extended attribute operations, and implement the XDR encode / decode logic for these operations. Add the user extended attributes operations to nfsd4_ops. Signed-off-by: Frank van der Linden <fllinden@amazon.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-06-24 06:39:26 +08:00
static __be32
nfsd4_getxattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
union nfsd4_op_u *u)
{
struct nfsd4_getxattr *getxattr = &u->getxattr;
return nfsd_getxattr(rqstp, &cstate->current_fh,
getxattr->getxa_name, &getxattr->getxa_buf,
&getxattr->getxa_len);
}
static __be32
nfsd4_setxattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
union nfsd4_op_u *u)
{
struct nfsd4_setxattr *setxattr = &u->setxattr;
__be32 ret;
if (opens_in_grace(SVC_NET(rqstp)))
return nfserr_grace;
ret = nfsd_setxattr(rqstp, &cstate->current_fh, setxattr->setxa_name,
setxattr->setxa_buf, setxattr->setxa_len,
setxattr->setxa_flags);
if (!ret)
set_change_info(&setxattr->setxa_cinfo, &cstate->current_fh);
return ret;
}
static __be32
nfsd4_listxattrs(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
union nfsd4_op_u *u)
{
/*
* Get the entire list, then copy out only the user attributes
* in the encode function.
*/
return nfsd_listxattr(rqstp, &cstate->current_fh,
&u->listxattrs.lsxa_buf, &u->listxattrs.lsxa_len);
}
static __be32
nfsd4_removexattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
union nfsd4_op_u *u)
{
struct nfsd4_removexattr *removexattr = &u->removexattr;
__be32 ret;
if (opens_in_grace(SVC_NET(rqstp)))
return nfserr_grace;
ret = nfsd_removexattr(rqstp, &cstate->current_fh,
removexattr->rmxa_name);
if (!ret)
set_change_info(&removexattr->rmxa_cinfo, &cstate->current_fh);
return ret;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/*
* NULL call.
*/
[PATCH] fix svc_procfunc declaration svc_procfunc instances return __be32, not int Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:28:45 +08:00
static __be32
sunrpc: properly type pc_func callbacks Drop the argp and resp arguments as they can trivially be derived from the rqstp argument. With that all functions now have the same prototype, and we can remove the unsafe casting to svc_procfunc as well as the svc_procfunc typedef itself. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-08 23:35:49 +08:00
nfsd4_proc_null(struct svc_rqst *rqstp)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
NFSD: Hoist status code encoding into XDR encoder functions The original intent was presumably to reduce code duplication. The trade-off was: - No support for an NFSD proc function returning a non-success RPC accept_stat value. - No support for void NFS replies to non-NULL procedures. - Everyone pays for the deduplication with a few extra conditional branches in a hot path. In addition, nfsd_dispatch() leaves *statp uninitialized in the success path, unlike svc_generic_dispatch(). Address all of these problems by moving the logic for encoding the NFS status code into the NFS XDR encoders themselves. Then update the NFS .pc_func methods to return an RPC accept_stat value. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2020-10-03 03:52:44 +08:00
return rpc_success;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
nfsd: make all of the nfsd stats per-network namespace We have a global set of counters that we modify for all of the nfsd operations, but now that we're exposing these stats across all network namespaces we need to make the stats also be per-network namespace. We already have some caching stats that are per-network namespace, so move these definitions into the same counter and then adjust all the helpers and users of these stats to provide the appropriate nfsd_net struct so that the stats are maintained for the per-network namespace objects. Signed-off-by: Josef Bacik <josef@toxicpanda.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-01-26 23:39:47 +08:00
static inline void nfsd4_increment_op_stats(struct nfsd_net *nn, u32 opnum)
[PATCH] knfsd: nfsd4: add per-operation server stats Add an nfs4 operations count array to nfsd_stats structure. The count is incremented in nfsd4_proc_compound() where all the operations are handled by the nfsv4 server. This count of individual nfsv4 operations is also entered into /proc filesystem. Signed-off-by: Shankar Anand<shanand@novell.com> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-07-10 19:45:44 +08:00
{
if (opnum >= FIRST_NFS4_OP && opnum <= LAST_NFS4_OP)
nfsd: make all of the nfsd stats per-network namespace We have a global set of counters that we modify for all of the nfsd operations, but now that we're exposing these stats across all network namespaces we need to make the stats also be per-network namespace. We already have some caching stats that are per-network namespace, so move these definitions into the same counter and then adjust all the helpers and users of these stats to provide the appropriate nfsd_net struct so that the stats are maintained for the per-network namespace objects. Signed-off-by: Josef Bacik <josef@toxicpanda.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-01-26 23:39:47 +08:00
percpu_counter_inc(&nn->counter[NFSD_STATS_NFS4_OP(opnum)]);
[PATCH] knfsd: nfsd4: add per-operation server stats Add an nfs4 operations count array to nfsd_stats structure. The count is incremented in nfsd4_proc_compound() where all the operations are handled by the nfsv4 server. This count of individual nfsv4 operations is also entered into /proc filesystem. Signed-off-by: Shankar Anand<shanand@novell.com> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-07-10 19:45:44 +08:00
}
nfsd4: const-ify nfsd4_ops nfsd4_ops contains function pointers, and marking it as constant avoids it being able to be used as an attach vector for code injections. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 05:46:47 +08:00
static const struct nfsd4_operation nfsd4_ops[];
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
[NFSD] uninline nfsd4_op_name() There doesn't seem to be a compelling reason why nfsd4_op_name() is marked as "inline": It's only used in a dprintk(), and as long as it has only one caller non-ancient gcc versions anyway inline it automatically. This patch fixes the following compile error with gcc 3.4: ... CC fs/nfsd/nfs4proc.o nfs4proc.c: In function `nfsd4_proc_compound': nfs4proc.c:854: sorry, unimplemented: inlining failed in call to nfs4proc.c:897: sorry, unimplemented: called from here make[3]: *** [fs/nfsd/nfs4proc.o] Error 1 Reported-by: Adrian Bunk <bunk@kernel.org> Signed-off-by: Adrian Bunk <bunk@kernel.org> [ Also made it "const char *" - Linus] Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-08-09 00:26:42 +08:00
static const char *nfsd4_op_name(unsigned opnum);
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
nfsd41: enforce NFS4ERR_SEQUENCE_POS operation order rules for minorversion != 0 only. Signed-off-by: Andy Adamson<andros@netapp.com> [nfsd41: do not verify nfserr_sequence_pos for minorversion 0] Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:28:12 +08:00
/*
nfsd4: complete enforcement of 4.1 op ordering Enforce the rules about compound op ordering. Motivated by implementing RECLAIM_COMPLETE, for which the client is implicit in the current session, so it is important to ensure a succesful SEQUENCE proceeds the RECLAIM_COMPLETE. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2010-04-22 00:27:19 +08:00
* Enforce NFSv4.1 COMPOUND ordering rules:
nfsd41: enforce NFS4ERR_SEQUENCE_POS operation order rules for minorversion != 0 only. Signed-off-by: Andy Adamson<andros@netapp.com> [nfsd41: do not verify nfserr_sequence_pos for minorversion 0] Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:28:12 +08:00
*
nfsd4: complete enforcement of 4.1 op ordering Enforce the rules about compound op ordering. Motivated by implementing RECLAIM_COMPLETE, for which the client is implicit in the current session, so it is important to ensure a succesful SEQUENCE proceeds the RECLAIM_COMPLETE. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2010-04-22 00:27:19 +08:00
* Also note, enforced elsewhere:
* - SEQUENCE other than as first op results in
* NFS4ERR_SEQUENCE_POS. (Enforced in nfsd4_sequence().)
nfsd4: support BIND_CONN_TO_SESSION Basic xdr and processing for BIND_CONN_TO_SESSION. This adds a connection to the list of connections associated with a session. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2010-10-05 11:12:59 +08:00
* - BIND_CONN_TO_SESSION must be the only op in its compound.
* (Enforced in nfsd4_bind_conn_to_session().)
nfsd4: complete enforcement of 4.1 op ordering Enforce the rules about compound op ordering. Motivated by implementing RECLAIM_COMPLETE, for which the client is implicit in the current session, so it is important to ensure a succesful SEQUENCE proceeds the RECLAIM_COMPLETE. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2010-04-22 00:27:19 +08:00
* - DESTROY_SESSION must be the final operation in a compound, if
* sessionid's in SEQUENCE and DESTROY_SESSION are the same.
* (Enforced in nfsd4_destroy_session().)
nfsd41: enforce NFS4ERR_SEQUENCE_POS operation order rules for minorversion != 0 only. Signed-off-by: Andy Adamson<andros@netapp.com> [nfsd41: do not verify nfserr_sequence_pos for minorversion 0] Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:28:12 +08:00
*/
nfsd4: complete enforcement of 4.1 op ordering Enforce the rules about compound op ordering. Motivated by implementing RECLAIM_COMPLETE, for which the client is implicit in the current session, so it is important to ensure a succesful SEQUENCE proceeds the RECLAIM_COMPLETE. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2010-04-22 00:27:19 +08:00
static __be32 nfs41_check_op_ordering(struct nfsd4_compoundargs *args)
nfsd41: enforce NFS4ERR_SEQUENCE_POS operation order rules for minorversion != 0 only. Signed-off-by: Andy Adamson<andros@netapp.com> [nfsd41: do not verify nfserr_sequence_pos for minorversion 0] Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:28:12 +08:00
{
nfsd: clarify check_op_ordering Document a couple things that confused me on a recent reading. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-06-14 03:04:59 +08:00
struct nfsd4_op *first_op = &args->ops[0];
nfsd4: complete enforcement of 4.1 op ordering Enforce the rules about compound op ordering. Motivated by implementing RECLAIM_COMPLETE, for which the client is implicit in the current session, so it is important to ensure a succesful SEQUENCE proceeds the RECLAIM_COMPLETE. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2010-04-22 00:27:19 +08:00
/* These ordering requirements don't apply to NFSv4.0: */
if (args->minorversion == 0)
return nfs_ok;
/* This is weird, but OK, not our problem: */
if (args->opcnt == 0)
return nfs_ok;
nfsd: clarify check_op_ordering Document a couple things that confused me on a recent reading. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-06-14 03:04:59 +08:00
if (first_op->status == nfserr_op_illegal)
nfsd4: complete enforcement of 4.1 op ordering Enforce the rules about compound op ordering. Motivated by implementing RECLAIM_COMPLETE, for which the client is implicit in the current session, so it is important to ensure a succesful SEQUENCE proceeds the RECLAIM_COMPLETE. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2010-04-22 00:27:19 +08:00
return nfs_ok;
nfsd: clarify check_op_ordering Document a couple things that confused me on a recent reading. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-06-14 03:04:59 +08:00
if (!(nfsd4_ops[first_op->opnum].op_flags & ALLOWED_AS_FIRST_OP))
nfsd4: complete enforcement of 4.1 op ordering Enforce the rules about compound op ordering. Motivated by implementing RECLAIM_COMPLETE, for which the client is implicit in the current session, so it is important to ensure a succesful SEQUENCE proceeds the RECLAIM_COMPLETE. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2010-04-22 00:27:19 +08:00
return nfserr_op_not_in_session;
nfsd: clarify check_op_ordering Document a couple things that confused me on a recent reading. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-06-14 03:04:59 +08:00
if (first_op->opnum == OP_SEQUENCE)
nfsd4: complete enforcement of 4.1 op ordering Enforce the rules about compound op ordering. Motivated by implementing RECLAIM_COMPLETE, for which the client is implicit in the current session, so it is important to ensure a succesful SEQUENCE proceeds the RECLAIM_COMPLETE. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2010-04-22 00:27:19 +08:00
return nfs_ok;
nfsd: clarify check_op_ordering Document a couple things that confused me on a recent reading. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-06-14 03:04:59 +08:00
/*
* So first_op is something allowed outside a session, like
* EXCHANGE_ID; but then it has to be the only op in the
* compound:
*/
nfsd4: complete enforcement of 4.1 op ordering Enforce the rules about compound op ordering. Motivated by implementing RECLAIM_COMPLETE, for which the client is implicit in the current session, so it is important to ensure a succesful SEQUENCE proceeds the RECLAIM_COMPLETE. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2010-04-22 00:27:19 +08:00
if (args->opcnt != 1)
return nfserr_not_only_op;
return nfs_ok;
nfsd41: enforce NFS4ERR_SEQUENCE_POS operation order rules for minorversion != 0 only. Signed-off-by: Andy Adamson<andros@netapp.com> [nfsd41: do not verify nfserr_sequence_pos for minorversion 0] Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:28:12 +08:00
}
nfsd4: opdesc will be useful outside nfs4proc.c Trivial cleanup, no change in behavior. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-07-07 05:51:29 +08:00
const struct nfsd4_operation *OPDESC(struct nfsd4_op *op)
nfsd4: introduce OPDESC helper Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-04-09 23:23:24 +08:00
{
return &nfsd4_ops[op->opnum];
}
nfsd: turn on reply cache for NFSv4 It's sort of ridiculous that we've never had a working reply cache for NFSv4. On the other hand, we may still not: our current reply cache is likely not very good, especially in the TCP case (which is the only case that matters for v4). What we really need here is some serious testing. Anyway, here's a start. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-01-25 01:11:02 +08:00
bool nfsd4_cache_this_op(struct nfsd4_op *op)
{
nfsd4: decoding errors can still be cached and require space Currently a non-idempotent op reply may be cached if it fails in the proc code but not if it fails at xdr decoding. I doubt there are any xdr-decoding-time errors that would make this a problem in practice, so this probably isn't a serious bug. The space estimates should also take into account space required for encoding of error returns. Again, not a practical problem, though it would become one after future patches which will tighten the space estimates. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-05-20 00:27:11 +08:00
if (op->opnum == OP_ILLEGAL)
return false;
nfsd4: make op_cacheresult another flag I'm not sure why I used a new field for this originally. Also, the differences between some of these flags are a little subtle; add some comments to explain. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-09-20 20:49:51 +08:00
return OPDESC(op)->op_flags & OP_CACHEME;
nfsd: turn on reply cache for NFSv4 It's sort of ridiculous that we've never had a working reply cache for NFSv4. On the other hand, we may still not: our current reply cache is likely not very good, especially in the TCP case (which is the only case that matters for v4). What we really need here is some serious testing. Anyway, here's a start. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-01-25 01:11:02 +08:00
}
nfsd4: fix wrongsec handling for PUTFH + op cases When PUTFH is followed by an operation that uses the filehandle, and when the current client is using a security flavor that is inconsistent with the given filehandle, we have a choice: we can return WRONGSEC either when the current filehandle is set using the PUTFH, or when the filehandle is first used by the following operation. Follow the recommendations of RFC 5661 in making this choice. (Our current behavior prevented the client from doing security negotiation by returning WRONGSEC on PUTFH+SECINFO_NO_NAME.) Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-04-09 05:00:50 +08:00
static bool need_wrongsec_check(struct svc_rqst *rqstp)
{
struct nfsd4_compoundres *resp = rqstp->rq_resp;
struct nfsd4_compoundargs *argp = rqstp->rq_argp;
struct nfsd4_op *this = &argp->ops[resp->opcnt - 1];
struct nfsd4_op *next = &argp->ops[resp->opcnt];
nfsd4: const-ify nfsd4_ops nfsd4_ops contains function pointers, and marking it as constant avoids it being able to be used as an attach vector for code injections. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 05:46:47 +08:00
const struct nfsd4_operation *thisd = OPDESC(this);
const struct nfsd4_operation *nextd;
nfsd4: fix wrongsec handling for PUTFH + op cases When PUTFH is followed by an operation that uses the filehandle, and when the current client is using a security flavor that is inconsistent with the given filehandle, we have a choice: we can return WRONGSEC either when the current filehandle is set using the PUTFH, or when the filehandle is first used by the following operation. Follow the recommendations of RFC 5661 in making this choice. (Our current behavior prevented the client from doing security negotiation by returning WRONGSEC on PUTFH+SECINFO_NO_NAME.) Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-04-09 05:00:50 +08:00
/*
* Most ops check wronsec on our own; only the putfh-like ops
* have special rules.
*/
if (!(thisd->op_flags & OP_IS_PUTFH_LIKE))
return false;
/*
* rfc 5661 2.6.3.1.1.6: don't bother erroring out a
* put-filehandle operation if we're not going to use the
* result:
*/
if (argp->opcnt == resp->opcnt)
return false;
nfsd4: fix crash on unknown operation number Unknown operation numbers are caught in nfsd4_decode_compound() which sets op->opnum to OP_ILLEGAL and op->status to nfserr_op_illegal. The error causes the main loop in nfsd4_proc_compound() to skip most processing. But nfsd4_proc_compound also peeks ahead at the next operation in one case and doesn't take similar precautions there. Cc: stable@vger.kernel.org Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-10-23 02:46:29 +08:00
if (next->opnum == OP_ILLEGAL)
return false;
nfsd4: fix wrongsec handling for PUTFH + op cases When PUTFH is followed by an operation that uses the filehandle, and when the current client is using a security flavor that is inconsistent with the given filehandle, we have a choice: we can return WRONGSEC either when the current filehandle is set using the PUTFH, or when the filehandle is first used by the following operation. Follow the recommendations of RFC 5661 in making this choice. (Our current behavior prevented the client from doing security negotiation by returning WRONGSEC on PUTFH+SECINFO_NO_NAME.) Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-04-09 05:00:50 +08:00
nextd = OPDESC(next);
/*
* Rest of 2.6.3.1.1: certain operations will return WRONGSEC
* errors themselves as necessary; others should check for them
* now:
*/
return !(nextd->op_flags & OP_HANDLES_WRONGSEC);
}
NFSD: allow inter server COPY to have a STALE source server fh The inter server to server COPY source server filehandle is a foreign filehandle as the COPY is sent to the destination server. Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-07 22:56:48 +08:00
#ifdef CONFIG_NFSD_V4_2_INTER_SSC
static void
check_if_stalefh_allowed(struct nfsd4_compoundargs *args)
{
struct nfsd4_op *op, *current_op = NULL, *saved_op = NULL;
struct nfsd4_copy *copy;
struct nfsd4_putfh *putfh;
int i;
/* traverse all operation and if it's a COPY compound, mark the
* source filehandle to skip verification
*/
for (i = 0; i < args->opcnt; i++) {
op = &args->ops[i];
if (op->opnum == OP_PUTFH)
current_op = op;
else if (op->opnum == OP_SAVEFH)
saved_op = current_op;
else if (op->opnum == OP_RESTOREFH)
current_op = saved_op;
else if (op->opnum == OP_COPY) {
copy = (struct nfsd4_copy *)&op->u;
if (!saved_op) {
op->status = nfserr_nofilehandle;
return;
}
putfh = (struct nfsd4_putfh *)&saved_op->u;
NFSD: Replace boolean fields in struct nfsd4_copy Clean up: saves 8 bytes, and we can replace check_and_set_stop_copy() with an atomic bitop. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-07-28 02:40:41 +08:00
if (nfsd4_ssc_is_inter(copy))
NFSD: allow inter server COPY to have a STALE source server fh The inter server to server COPY source server filehandle is a foreign filehandle as the COPY is sent to the destination server. Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-07 22:56:48 +08:00
putfh->no_verify = true;
}
}
}
#else
static void
check_if_stalefh_allowed(struct nfsd4_compoundargs *args)
{
}
#endif
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/*
* COMPOUND call.
*/
[PATCH] fix svc_procfunc declaration svc_procfunc instances return __be32, not int Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:28:45 +08:00
static __be32
sunrpc: properly type pc_func callbacks Drop the argp and resp arguments as they can trivially be derived from the rqstp argument. With that all functions now have the same prototype, and we can remove the unsafe casting to svc_procfunc as well as the svc_procfunc typedef itself. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-08 23:35:49 +08:00
nfsd4_proc_compound(struct svc_rqst *rqstp)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
sunrpc: properly type pc_func callbacks Drop the argp and resp arguments as they can trivially be derived from the rqstp argument. With that all functions now have the same prototype, and we can remove the unsafe casting to svc_procfunc as well as the svc_procfunc typedef itself. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-08 23:35:49 +08:00
struct nfsd4_compoundargs *args = rqstp->rq_argp;
struct nfsd4_compoundres *resp = rqstp->rq_resp;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
struct nfsd4_op *op;
nfsd: embed nfsd4_current_state in nfsd4_compoundres Remove the allocation of struct nfsd4_compound_state. Signed-off-by: Andy Adamson <andros@netapp.com> Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-03-28 16:30:52 +08:00
struct nfsd4_compound_state *cstate = &resp->cstate;
NFSD: simplify saved/current fh uses in nfsd4_proc_compound Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-26 17:10:37 +08:00
struct svc_fh *current_fh = &cstate->current_fh;
struct svc_fh *save_fh = &cstate->save_fh;
nfsd: Allow containers to set supported nfs versions Support use of the --nfs-version/--no-nfs-version arguments to rpc.nfsd in containers. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-04-09 23:46:19 +08:00
struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
[PATCH] nfsd: nfs4 code returns error values in net-endian Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:28:59 +08:00
__be32 status;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
NFSD: Extract the svcxdr_init_encode() helper NFSD initializes an encode xdr_stream only after the RPC layer has already inserted the RPC Reply header. Thus it behaves differently than xdr_init_encode does, which assumes the passed-in xdr_buf is entirely devoid of content. nfs4proc.c has this server-side stream initialization helper, but it is visible only to the NFSv4 code. Move this helper to a place that can be accessed by NFSv2 and NFSv3 server XDR functions. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-10-28 03:53:42 +08:00
resp->xdr = &rqstp->rq_res_stream;
NFSD: Save location of NFSv4 COMPOUND status Refactor: Currently nfs4svc_encode_compoundres() relies on the NFS dispatcher to pass in the buffer location of the COMPOUND status. Instead, save that buffer location in struct nfsd4_compoundres. The compound tag follows immediately after. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-10-13 22:40:59 +08:00
resp->statusp = resp->xdr->p;
NFSD: Extract the svcxdr_init_encode() helper NFSD initializes an encode xdr_stream only after the RPC layer has already inserted the RPC Reply header. Thus it behaves differently than xdr_init_encode does, which assumes the passed-in xdr_buf is entirely devoid of content. nfs4proc.c has this server-side stream initialization helper, but it is visible only to the NFSv4 code. Move this helper to a place that can be accessed by NFSv2 and NFSv3 server XDR functions. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-10-28 03:53:42 +08:00
/* reserve space for: NFS status code */
xdr_reserve_space(resp->xdr, XDR_UNIT);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* reserve space for: taglen, tag, and opcnt */
NFSD: Extract the svcxdr_init_encode() helper NFSD initializes an encode xdr_stream only after the RPC layer has already inserted the RPC Reply header. Thus it behaves differently than xdr_init_encode does, which assumes the passed-in xdr_buf is entirely devoid of content. nfs4proc.c has this server-side stream initialization helper, but it is visible only to the NFSv4 code. Move this helper to a place that can be accessed by NFSv2 and NFSv3 server XDR functions. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-10-28 03:53:42 +08:00
xdr_reserve_space(resp->xdr, XDR_UNIT * 2 + args->taglen);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
resp->taglen = args->taglen;
resp->tag = args->tag;
resp->rqstp = rqstp;
NFSD: simplify saved/current fh uses in nfsd4_proc_compound Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-26 17:10:37 +08:00
cstate->minorversion = args->minorversion;
fh_init(current_fh, NFS4_FHSIZE);
fh_init(save_fh, NFS4_FHSIZE);
nfsd: disable deferral for NFSv4 Now that a slight delay in getting a reply to an upcall doesn't require deferring of requests, request deferral for all NFSv4 requests - the concept doesn't really fit with the v4 model. Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2010-08-12 15:04:07 +08:00
/*
* Don't use the deferral mechanism for NFSv4; compounds make it
* too hard to avoid non-idempotency problems.
*/
Revert "SUNRPC: Use RMW bitops in single-threaded hot paths" The premise that "Once an svc thread is scheduled and executing an RPC, no other processes will touch svc_rqst::rq_flags" is false. svc_xprt_enqueue() examines the RQ_BUSY flag in scheduled nfsd threads when determining which thread to wake up next. Found via KCSAN. Fixes: 28df0988815f ("SUNRPC: Use RMW bitops in single-threaded hot paths") Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-01-07 01:43:37 +08:00
clear_bit(RQ_USEDEFERRAL, &rqstp->rq_flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/*
* According to RFC3010, this takes precedence over all other errors.
*/
status = nfserr_minor_vers_mismatch;
nfsd: Allow containers to set supported nfs versions Support use of the --nfs-version/--no-nfs-version arguments to rpc.nfsd in containers. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-04-09 23:46:19 +08:00
if (nfsd_minorversion(nn, args->minorversion, NFSD_TEST) <= 0)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
goto out;
nfsd4: complete enforcement of 4.1 op ordering Enforce the rules about compound op ordering. Motivated by implementing RECLAIM_COMPLETE, for which the client is implicit in the current session, so it is important to ensure a succesful SEQUENCE proceeds the RECLAIM_COMPLETE. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2010-04-22 00:27:19 +08:00
status = nfs41_check_op_ordering(args);
if (status) {
nfsd41: enforce NFS4ERR_SEQUENCE_POS operation order rules for minorversion != 0 only. Signed-off-by: Andy Adamson<andros@netapp.com> [nfsd41: do not verify nfserr_sequence_pos for minorversion 0] Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:28:12 +08:00
op = &args->ops[0];
nfsd4: complete enforcement of 4.1 op ordering Enforce the rules about compound op ordering. Motivated by implementing RECLAIM_COMPLETE, for which the client is implicit in the current session, so it is important to ensure a succesful SEQUENCE proceeds the RECLAIM_COMPLETE. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2010-04-22 00:27:19 +08:00
op->status = status;
nfsd: fix corrupted reply to badly ordered compound We're encoding a single op in the reply but leaving the number of ops zero, so the reply makes no sense. Somewhat academic as this isn't a case any real client will hit, though in theory perhaps that could change in a future protocol extension. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-06-14 03:21:35 +08:00
resp->opcnt = 1;
nfsd41: enforce NFS4ERR_SEQUENCE_POS operation order rules for minorversion != 0 only. Signed-off-by: Andy Adamson<andros@netapp.com> [nfsd41: do not verify nfserr_sequence_pos for minorversion 0] Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:28:12 +08:00
goto encode_op;
}
NFSD: allow inter server COPY to have a STALE source server fh The inter server to server COPY source server filehandle is a foreign filehandle as the COPY is sent to the destination server. Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-07 22:56:48 +08:00
check_if_stalefh_allowed(args);
nfsd41: enforce NFS4ERR_SEQUENCE_POS operation order rules for minorversion != 0 only. Signed-off-by: Andy Adamson<andros@netapp.com> [nfsd41: do not verify nfserr_sequence_pos for minorversion 0] Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:28:12 +08:00
nfsd: clients don't need to break their own delegations We currently revoke read delegations on any write open or any operation that modifies file data or metadata (including rename, link, and unlink). But if the delegation in question is the only read delegation and is held by the client performing the operation, that's not really necessary. It's not always possible to prevent this in the NFSv4.0 case, because there's not always a way to determine which client an NFSv4.0 delegation came from. (In theory we could try to guess this from the transport layer, e.g., by assuming all traffic on a given TCP connection comes from the same client. But that's not really correct.) In the NFSv4.1 case the session layer always tells us the client. This patch should remove such self-conflicts in all cases where we can reliably determine the client from the compound. To do that we need to track "who" is performing a given (possibly lease-breaking) file operation. We're doing that by storing the information in the svc_rqst and using kthread_data() to map the current task back to a svc_rqst. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-07-29 04:35:15 +08:00
rqstp->rq_lease_breaker = (void **)&cstate->clp;
NFSD: Trace NFSv4 COMPOUND tags The Linux NFSv4 client implementation does not use COMPOUND tags, but the Solaris and MacOS implementations do, and so does pynfs. Record these eye-catchers in the server's trace buffer to annotate client requests while troubleshooting. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Reviewed-by: Jeff Layton <jlayton@kernel.org>
2022-09-09 06:13:48 +08:00
trace_nfsd_compound(rqstp, args->tag, args->taglen, args->client_opcnt);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
while (!status && resp->opcnt < args->opcnt) {
op = &args->ops[resp->opcnt++];
NFSD: Fix handling of oversized NFSv4 COMPOUND requests If an NFS server returns NFS4ERR_RESOURCE on the first operation in an NFSv4 COMPOUND, there's no way for a client to know where the problem is and then simplify the compound to make forward progress. So instead, make NFSD process as many operations in an oversized COMPOUND as it can and then return NFS4ERR_RESOURCE on the first operation it did not process. pynfs NFSv4.0 COMP6 exercises this case, but checks only for the COMPOUND status code, not whether the server has processed any of the operations. pynfs NFSv4.1 SEQ6 and SEQ7 exercise the NFSv4.1 case, which detects too many operations per COMPOUND by checking against the limits negotiated when the session was created. Suggested-by: Bruce Fields <bfields@fieldses.org> Fixes: 0078117c6d91 ("nfsd: return RESOURCE not GARBAGE_ARGS on too many ops") Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-06 03:33:32 +08:00
if (unlikely(resp->opcnt == NFSD_MAX_OPS_PER_COMPOUND)) {
/* If there are still more operations to process,
* stop here and report NFS4ERR_RESOURCE. */
if (cstate->minorversion == 0 &&
args->client_opcnt > resp->opcnt) {
op->status = nfserr_resource;
goto encode_op;
}
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/*
* The XDR decode routines may have pre-set op->status;
* for example, if there is a miscellaneous XDR error
* it will be set to nfserr_bad_xdr.
*/
nfsd4: handle seqid-mutating open errors from xdr decoding If a client sets an owner (or group_owner or acl) attribute on open for create, and the mapping of that owner to an id fails, then we return BAD_OWNER. But BAD_OWNER is a seqid-mutating error, so we can't shortcut the open processing that case: we have to at least look up the owner so we can find the seqid to bump. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-03-01 04:51:49 +08:00
if (op->status) {
if (op->opnum == OP_OPEN)
op->status = nfsd4_open_omfg(rqstp, cstate, op);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
goto encode_op;
nfsd4: handle seqid-mutating open errors from xdr decoding If a client sets an owner (or group_owner or acl) attribute on open for create, and the mapping of that owner to an id fails, then we return BAD_OWNER. But BAD_OWNER is a seqid-mutating error, so we can't shortcut the open processing that case: we have to at least look up the owner so we can find the seqid to bump. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-03-01 04:51:49 +08:00
}
NFSD: allow inter server COPY to have a STALE source server fh The inter server to server COPY source server filehandle is a foreign filehandle as the COPY is sent to the destination server. Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-07 22:56:48 +08:00
if (!current_fh->fh_dentry &&
!HAS_FH_FLAG(current_fh, NFSD4_FH_FOREIGN)) {
nfsd4: opdesc will be useful outside nfs4proc.c Trivial cleanup, no change in behavior. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-07-07 05:51:29 +08:00
if (!(op->opdesc->op_flags & ALLOWED_WITHOUT_FH)) {
[PATCH] knfsd: nfsd4: actually use all the pieces to implement referrals Use all the pieces set up so far to implement referral support, allowing return of NFS4ERR_MOVED and fs_locations attribute. Signed-off-by: Manoj Naik <manoj@almaden.ibm.com> Signed-off-by: Fred Isaman <iisaman@citi.umich.edu> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 17:16:20 +08:00
op->status = nfserr_nofilehandle;
goto encode_op;
}
NFSD: allow inter server COPY to have a STALE source server fh The inter server to server COPY source server filehandle is a foreign filehandle as the COPY is sent to the destination server. Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-10-07 22:56:48 +08:00
} else if (current_fh->fh_export &&
current_fh->fh_export->ex_fslocs.migrated &&
nfsd4: opdesc will be useful outside nfs4proc.c Trivial cleanup, no change in behavior. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-07-07 05:51:29 +08:00
!(op->opdesc->op_flags & ALLOWED_ON_ABSENT_FS)) {
[PATCH] knfsd: nfsd4: actually use all the pieces to implement referrals Use all the pieces set up so far to implement referral support, allowing return of NFS4ERR_MOVED and fs_locations attribute. Signed-off-by: Manoj Naik <manoj@almaden.ibm.com> Signed-off-by: Fred Isaman <iisaman@citi.umich.edu> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 17:16:20 +08:00
op->status = nfserr_moved;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
goto encode_op;
}
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
NFSD: Move fill_pre_wcc() and fill_post_wcc() These functions are related to file handle processing and have nothing to do with XDR encoding or decoding. Also they are no longer NFSv3-specific. As a clean-up, move their definitions to a more appropriate location. WCC is also an NFSv3-specific term, so rename them as general-purpose helpers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2021-12-25 03:36:49 +08:00
fh_clear_pre_post_attrs(current_fh);
NFSD: Clear wcc data between compound ops Testing NFS4.0 by pynfs, I got some messeages as, "nfsd: inode locked twice during operation." When one compound RPC contains two or more ops that locks the filehandle,the second op will cause the message. As two SETATTR ops, after the first SETATTR, nfsd will not call fh_put() to release current filehandle, it means filehandle have unlocked with fh_post_saved = 1. The second SETATTR find fh_post_saved = 1, and printk the message. v2: introduce helper fh_clear_wcc(). Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-29 10:23:47 +08:00
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
/* If op is non-idempotent */
nfsd4: opdesc will be useful outside nfs4proc.c Trivial cleanup, no change in behavior. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-07-07 05:51:29 +08:00
if (op->opdesc->op_flags & OP_MODIFIES_SOMETHING) {
nfsd4: session needs room for following op to error out Cc: stable@vger.kernel.org Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-01-29 05:01:04 +08:00
/*
nfsd4: size-checking cleanup Better variable name, some comments, etc. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-12 05:54:34 +08:00
* Don't execute this op if we couldn't encode a
NFSD: Fix space and spelling mistake Add a blank space after ','. Change 'succesful' to 'successful'. Signed-off-by: Zhang Jiaming <jiaming@nfschina.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-06-23 16:20:05 +08:00
* successful reply:
nfsd4: size-checking cleanup Better variable name, some comments, etc. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-12 05:54:34 +08:00
*/
nfsd4: opdesc will be useful outside nfs4proc.c Trivial cleanup, no change in behavior. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-07-07 05:51:29 +08:00
u32 plen = op->opdesc->op_rsize_bop(rqstp, op);
nfsd4: size-checking cleanup Better variable name, some comments, etc. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-12 05:54:34 +08:00
/*
* Plus if there's another operation, make sure
nfsd4: session needs room for following op to error out Cc: stable@vger.kernel.org Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-01-29 05:01:04 +08:00
* we'll have space to at least encode an error:
*/
if (resp->opcnt < args->opcnt)
plen += COMPOUND_ERR_SLACK_SPACE;
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
op->status = nfsd4_check_resp_size(resp, plen);
}
if (op->status)
goto encode_op;
nfsd4: opdesc will be useful outside nfs4proc.c Trivial cleanup, no change in behavior. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-07-07 05:51:29 +08:00
if (op->opdesc->op_get_currentstateid)
op->opdesc->op_get_currentstateid(cstate, &op->u);
op->status = op->opdesc->op_func(rqstp, cstate, &op->u);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
nfsd4: fix null dereference on replay if we receive a compound such that: - the sessionid, slot, and sequence number in the SEQUENCE op match a cached succesful reply with N ops, and - the Nth operation of the compound is a PUTFH, PUTPUBFH, PUTROOTFH, or RESTOREFH, then nfsd4_sequence will return 0 and set cstate->status to nfserr_replay_cache. The current filehandle will not be set. This will cause us to call check_nfsd_access with first argument NULL. To nfsd4_compound it looks like we just succesfully executed an operation that set a filehandle, but the current filehandle is not set. Fix this by moving the nfserr_replay_cache earlier. There was never any reason to have it after the encode_op label, since the only case where he hit that is when opdesc->op_func sets it. Note that there are two ways we could hit this case: - a client is resending a previously sent compound that ended with one of the four PUTFH-like operations, or - a client is sending a *new* compound that (incorrectly) shares sessionid, slot, and sequence number with a previously sent compound, and the length of the previously sent compound happens to match the position of a PUTFH-like operation in the new compound. The second is obviously incorrect client behavior. The first is also very strange--the only purpose of a PUTFH-like operation is to set the current filehandle to be used by the following operation, so there's no point in having it as the last in a compound. So it's likely this requires a buggy or malicious client to reproduce. Reported-by: Scott Mayhew <smayhew@redhat.com> Cc: stable@kernel.vger.org Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-05-24 00:24:40 +08:00
/* Only from SEQUENCE */
if (cstate->status == nfserr_replay_cache) {
dprintk("%s NFS4.1 replay from cache\n", __func__);
status = op->status;
goto out;
}
nfsd41: handle current stateid in open and close Signed-off-by: Tigran Mkrtchyan <kofemann@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-02-14 05:55:24 +08:00
if (!op->status) {
nfsd4: opdesc will be useful outside nfs4proc.c Trivial cleanup, no change in behavior. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-07-07 05:51:29 +08:00
if (op->opdesc->op_set_currentstateid)
op->opdesc->op_set_currentstateid(cstate, &op->u);
nfsd41: handle current stateid in open and close Signed-off-by: Tigran Mkrtchyan <kofemann@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-02-14 05:55:24 +08:00
nfsd4: opdesc will be useful outside nfs4proc.c Trivial cleanup, no change in behavior. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-07-07 05:51:29 +08:00
if (op->opdesc->op_flags & OP_CLEAR_STATEID)
nfsd41: use current stateid by value Signed-off-by: Tigran Mkrtchyan <kofemann@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-02-14 05:55:32 +08:00
clear_current_stateid(cstate);
nfsd41: handle current stateid in open and close Signed-off-by: Tigran Mkrtchyan <kofemann@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-02-14 05:55:24 +08:00
nfsd4: avoid NULL deference on strange COPY compounds With cross-server COPY we've introduced the possibility that the current or saved filehandle might not have fh_dentry/fh_export filled in, but we missed a place that assumed it was. I think this could be triggered by a compound like: PUTFH(foreign filehandle) GETATTR SAVEFH COPY First, check_if_stalefh_allowed sets no_verify on the first (PUTFH) op. Then op_func = nfsd4_putfh runs and leaves current_fh->fh_export NULL. need_wrongsec_check returns true, since this PUTFH has OP_IS_PUTFH_LIKE set and GETATTR does not have OP_HANDLES_WRONGSEC set. We should probably also consider tightening the checks in check_if_stalefh_allowed and double-checking that we don't assume the filehandle is verified elsewhere in the compound. But I think this fixes the immediate issue. Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Fixes: 4e48f1cccab3 "NFSD: allow inter server COPY to have... " Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2019-12-07 05:07:32 +08:00
if (current_fh->fh_export &&
need_wrongsec_check(rqstp))
NFSD: simplify saved/current fh uses in nfsd4_proc_compound Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-26 17:10:37 +08:00
op->status = check_nfsd_access(current_fh->fh_export, rqstp);
nfsd41: handle current stateid in open and close Signed-off-by: Tigran Mkrtchyan <kofemann@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-02-14 05:55:24 +08:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
encode_op:
[PATCH] nfsd: nfs_replay_me We are using NFS_REPLAY_ME as a special error value that is never leaked to clients. That works fine; the only problem is mixing host- and network- endian values in the same objects. Network-endian equivalent would work just as fine; switch to it. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-20 14:29:03 +08:00
if (op->status == nfserr_replay_me) {
[PATCH] knfsd: nfsd4: move replay_owner to cstate Tuck away the replay_owner in the cstate while we're at it. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:28 +08:00
op->replay = &cstate->replay_owner->so_replay;
NFSD: Extract the svcxdr_init_encode() helper NFSD initializes an encode xdr_stream only after the RPC layer has already inserted the RPC Reply header. Thus it behaves differently than xdr_init_encode does, which assumes the passed-in xdr_buf is entirely devoid of content. nfs4proc.c has this server-side stream initialization helper, but it is visible only to the NFSv4 code. Move this helper to a place that can be accessed by NFSv2 and NFSv3 server XDR functions. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-10-28 03:53:42 +08:00
nfsd4_encode_replay(resp->xdr, op);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
status = op->status = op->replay->rp_status;
} else {
nfsd4_encode_operation(resp, op);
status = op->status;
}
nfsd: dprint each op status in nfsd4_proc_compound Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-12-16 01:40:49 +08:00
NFSD: Fix handling of oversized NFSv4 COMPOUND requests If an NFS server returns NFS4ERR_RESOURCE on the first operation in an NFSv4 COMPOUND, there's no way for a client to know where the problem is and then simplify the compound to make forward progress. So instead, make NFSD process as many operations in an oversized COMPOUND as it can and then return NFS4ERR_RESOURCE on the first operation it did not process. pynfs NFSv4.0 COMP6 exercises this case, but checks only for the COMPOUND status code, not whether the server has processed any of the operations. pynfs NFSv4.1 SEQ6 and SEQ7 exercise the NFSv4.1 case, which detects too many operations per COMPOUND by checking against the limits negotiated when the session was created. Suggested-by: Bruce Fields <bfields@fieldses.org> Fixes: 0078117c6d91 ("nfsd: return RESOURCE not GARBAGE_ARGS on too many ops") Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-06 03:33:32 +08:00
trace_nfsd_compound_status(args->client_opcnt, resp->opcnt,
status, nfsd4_op_name(op->opnum));
nfsd: dprint each op status in nfsd4_proc_compound Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-12-16 01:40:49 +08:00
nfsd: Add a mutex to protect the NFSv4.0 open owner replay cache We don't want to rely on the client_mutex for protection in the case of NFSv4 open owners. Instead, we add a mutex that will only be taken for NFSv4.0 state mutating operations, and that will be released once the entire compound is done. Also, ensure that nfsd4_cstate_assign_replay/nfsd4_cstate_clear_replay take a reference to the stateowner when they are using it for NFSv4.0 open and lock replay caching. Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Jeff Layton <jlayton@primarydata.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-07-30 09:34:27 +08:00
nfsd4_cstate_clear_replay(cstate);
nfsd: make all of the nfsd stats per-network namespace We have a global set of counters that we modify for all of the nfsd operations, but now that we're exposing these stats across all network namespaces we need to make the stats also be per-network namespace. We already have some caching stats that are per-network namespace, so move these definitions into the same counter and then adjust all the helpers and users of these stats to provide the appropriate nfsd_net struct so that the stats are maintained for the per-network namespace objects. Signed-off-by: Josef Bacik <josef@toxicpanda.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-01-26 23:39:47 +08:00
nfsd4_increment_op_stats(nn, op->opnum);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
NFSD: simplify saved/current fh uses in nfsd4_proc_compound Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-26 17:10:37 +08:00
fh_put(current_fh);
fh_put(save_fh);
BUG_ON(cstate->replay_owner);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
out:
NFSD: Hoist status code encoding into XDR encoder functions The original intent was presumably to reduce code duplication. The trade-off was: - No support for an NFSD proc function returning a non-success RPC accept_stat value. - No support for void NFS replies to non-NULL procedures. - Everyone pays for the deduplication with a few extra conditional branches in a hot path. In addition, nfsd_dispatch() leaves *statp uninitialized in the success path, unlike svc_generic_dispatch(). Address all of these problems by moving the logic for encoding the NFS status code into the NFS XDR encoders themselves. Then update the NFS .pc_func methods to return an RPC accept_stat value. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2020-10-03 03:52:44 +08:00
cstate->status = status;
nfsd: don't use the deferral service, return NFS4ERR_DELAY On an NFSv4.1 server cache miss that causes an upcall, NFS4ERR_DELAY will be returned. It is up to the NFSv4.1 client to resend only the operations that have not been processed. Initialize rq_usedeferral to 1 in svc_process(). It sill be turned off in nfsd4_proc_compound() only when NFSv4.1 Sessions are used. Note: this isn't an adequate solution on its own. It's acceptable as a way to get some minimal 4.1 up and working, but we're going to have to find a way to avoid returning DELAY in all common cases before 4.1 can really be considered ready. Signed-off-by: Andy Adamson <andros@netapp.com> Signed-off-by: Benny Halevy <bhalevy@panasas.com> [nfsd41: reverse rq_nodeferral negative logic] Signed-off-by: Benny Halevy <bhalevy@panasas.com> [sunrpc: initialize rq_usedeferral] Signed-off-by: Andy Adamson <andros@netapp.com> Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:27:32 +08:00
/* Reset deferral mechanism for RPC deferrals */
Revert "SUNRPC: Use RMW bitops in single-threaded hot paths" The premise that "Once an svc thread is scheduled and executing an RPC, no other processes will touch svc_rqst::rq_flags" is false. svc_xprt_enqueue() examines the RQ_BUSY flag in scheduled nfsd threads when determining which thread to wake up next. Found via KCSAN. Fixes: 28df0988815f ("SUNRPC: Use RMW bitops in single-threaded hot paths") Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-01-07 01:43:37 +08:00
set_bit(RQ_USEDEFERRAL, &rqstp->rq_flags);
NFSD: Hoist status code encoding into XDR encoder functions The original intent was presumably to reduce code duplication. The trade-off was: - No support for an NFSD proc function returning a non-success RPC accept_stat value. - No support for void NFS replies to non-NULL procedures. - Everyone pays for the deduplication with a few extra conditional branches in a hot path. In addition, nfsd_dispatch() leaves *statp uninitialized in the success path, unlike svc_generic_dispatch(). Address all of these problems by moving the logic for encoding the NFS status code into the NFS XDR encoders themselves. Then update the NFS .pc_func methods to return an RPC accept_stat value. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2020-10-03 03:52:44 +08:00
return rpc_success;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
#define op_encode_hdr_size (2)
#define op_encode_stateid_maxsz (XDR_QUADLEN(NFS4_STATEID_SIZE))
#define op_encode_verifier_maxsz (XDR_QUADLEN(NFS4_VERIFIER_SIZE))
#define op_encode_change_info_maxsz (5)
#define nfs4_fattr_bitmap_maxsz (4)
nfsd4: don't try to encode conflicting owner if low on space I ran into this corner case in testing: in theory clients can provide state owners up to 1024 bytes long. In the sessions case there might be a risk of this pushing us over the DRC slot size. The conflicting owner isn't really that important, so let's humor a client that provides a small maxresponsize_cached by allowing ourselves to return without the conflicting owner instead of outright failing the operation. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-11 00:19:10 +08:00
/* We'll fall back on returning no lockowner if run out of space: */
#define op_encode_lockowner_maxsz (0)
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
#define op_encode_lock_denied_maxsz (8 + op_encode_lockowner_maxsz)
#define nfs4_owner_maxsz (1 + XDR_QUADLEN(IDMAP_NAMESZ))
#define op_encode_ace_maxsz (3 + nfs4_owner_maxsz)
#define op_encode_delegation_maxsz (1 + op_encode_stateid_maxsz + 1 + \
op_encode_ace_maxsz)
#define op_encode_channel_attrs_maxsz (6 + 1 + 1)
NFSD: Cap rsize_bop result based on send buffer size Since before the git era, NFSD has conserved the number of pages held by each nfsd thread by combining the RPC receive and send buffers into a single array of pages. This works because there are no cases where an operation needs a large RPC Call message and a large RPC Reply at the same time. Once an RPC Call has been received, svc_process() updates svc_rqst::rq_res to describe the part of rq_pages that can be used for constructing the Reply. This means that the send buffer (rq_res) shrinks when the received RPC record containing the RPC Call is large. Add an NFSv4 helper that computes the size of the send buffer. It replaces svc_max_payload() in spots where svc_max_payload() returns a value that might be larger than the remaining send buffer space. Callers who need to know the transport's actual maximum payload size will continue to use svc_max_payload(). Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-02 03:29:55 +08:00
/*
* The _rsize() helpers are invoked by the NFSv4 COMPOUND decoder, which
* is called before sunrpc sets rq_res.buflen. Thus we have to compute
* the maximum payload size here, based on transport limits and the size
* of the remaining space in the rq_pages array.
*/
static u32 nfsd4_max_payload(const struct svc_rqst *rqstp)
{
u32 buflen;
buflen = (rqstp->rq_page_end - rqstp->rq_next_page) * PAGE_SIZE;
buflen -= rqstp->rq_auth_slack;
buflen -= rqstp->rq_res.head[0].iov_len;
return min_t(u32, buflen, svc_max_payload(rqstp));
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_only_status_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
{
return (op_encode_hdr_size) * sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_status_stateid_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
{
return (op_encode_hdr_size + op_encode_stateid_maxsz)* sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_access_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
NFSD: Get response size before operation for all RPCs NFSD usess PAGE_SIZE as the reply size estimate for RPCs which don't support op_rsize_bop(), A PAGE_SIZE (4096) is larger than many real response sizes, eg, access (op_encode_hdr_size + 2), seek (op_encode_hdr_size + 3). This patch just adds op_rsize_bop() for all RPCs getting response size. An overestimate is generally safe but the tighter estimates are probably better. Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-02-03 22:36:00 +08:00
{
/* ac_supported, ac_resp_access */
return (op_encode_hdr_size + 2)* sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_commit_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
{
return (op_encode_hdr_size + op_encode_verifier_maxsz) * sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_create_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
{
return (op_encode_hdr_size + op_encode_change_info_maxsz
+ nfs4_fattr_bitmap_maxsz) * sizeof(__be32);
}
nfsd4: better estimate of getattr response size We plan to use this estimate to decide whether or not to allow zero-copy reads. Currently we're assuming all getattr's are a page, which can be both too small (ACLs e.g. may be arbitrarily long) and too large (after an upcoming read patch this will unnecessarily prevent zero copy reads in any read compound also containing a getattr). Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-24 00:01:48 +08:00
/*
* Note since this is an idempotent operation we won't insist on failing
* the op prematurely if the estimate is too large. We may turn off splice
* reads unnecessarily.
*/
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_getattr_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd4: better estimate of getattr response size We plan to use this estimate to decide whether or not to allow zero-copy reads. Currently we're assuming all getattr's are a page, which can be both too small (ACLs e.g. may be arbitrarily long) and too large (after an upcoming read patch this will unnecessarily prevent zero copy reads in any read compound also containing a getattr). Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-24 00:01:48 +08:00
{
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
const u32 *bmap = op->u.getattr.ga_bmval;
nfsd4: better estimate of getattr response size We plan to use this estimate to decide whether or not to allow zero-copy reads. Currently we're assuming all getattr's are a page, which can be both too small (ACLs e.g. may be arbitrarily long) and too large (after an upcoming read patch this will unnecessarily prevent zero copy reads in any read compound also containing a getattr). Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-24 00:01:48 +08:00
u32 bmap0 = bmap[0], bmap1 = bmap[1], bmap2 = bmap[2];
u32 ret = 0;
if (bmap0 & FATTR4_WORD0_ACL)
NFSD: Cap rsize_bop result based on send buffer size Since before the git era, NFSD has conserved the number of pages held by each nfsd thread by combining the RPC receive and send buffers into a single array of pages. This works because there are no cases where an operation needs a large RPC Call message and a large RPC Reply at the same time. Once an RPC Call has been received, svc_process() updates svc_rqst::rq_res to describe the part of rq_pages that can be used for constructing the Reply. This means that the send buffer (rq_res) shrinks when the received RPC record containing the RPC Call is large. Add an NFSv4 helper that computes the size of the send buffer. It replaces svc_max_payload() in spots where svc_max_payload() returns a value that might be larger than the remaining send buffer space. Callers who need to know the transport's actual maximum payload size will continue to use svc_max_payload(). Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-02 03:29:55 +08:00
return nfsd4_max_payload(rqstp);
nfsd4: better estimate of getattr response size We plan to use this estimate to decide whether or not to allow zero-copy reads. Currently we're assuming all getattr's are a page, which can be both too small (ACLs e.g. may be arbitrarily long) and too large (after an upcoming read patch this will unnecessarily prevent zero copy reads in any read compound also containing a getattr). Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-24 00:01:48 +08:00
if (bmap0 & FATTR4_WORD0_FS_LOCATIONS)
NFSD: Cap rsize_bop result based on send buffer size Since before the git era, NFSD has conserved the number of pages held by each nfsd thread by combining the RPC receive and send buffers into a single array of pages. This works because there are no cases where an operation needs a large RPC Call message and a large RPC Reply at the same time. Once an RPC Call has been received, svc_process() updates svc_rqst::rq_res to describe the part of rq_pages that can be used for constructing the Reply. This means that the send buffer (rq_res) shrinks when the received RPC record containing the RPC Call is large. Add an NFSv4 helper that computes the size of the send buffer. It replaces svc_max_payload() in spots where svc_max_payload() returns a value that might be larger than the remaining send buffer space. Callers who need to know the transport's actual maximum payload size will continue to use svc_max_payload(). Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-02 03:29:55 +08:00
return nfsd4_max_payload(rqstp);
nfsd4: better estimate of getattr response size We plan to use this estimate to decide whether or not to allow zero-copy reads. Currently we're assuming all getattr's are a page, which can be both too small (ACLs e.g. may be arbitrarily long) and too large (after an upcoming read patch this will unnecessarily prevent zero copy reads in any read compound also containing a getattr). Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-24 00:01:48 +08:00
if (bmap1 & FATTR4_WORD1_OWNER) {
ret += IDMAP_NAMESZ + 4;
bmap1 &= ~FATTR4_WORD1_OWNER;
}
if (bmap1 & FATTR4_WORD1_OWNER_GROUP) {
ret += IDMAP_NAMESZ + 4;
bmap1 &= ~FATTR4_WORD1_OWNER_GROUP;
}
if (bmap0 & FATTR4_WORD0_FILEHANDLE) {
ret += NFS4_FHSIZE + 4;
bmap0 &= ~FATTR4_WORD0_FILEHANDLE;
}
if (bmap2 & FATTR4_WORD2_SECURITY_LABEL) {
nfsd: Remove duplicate macro define for max sec label length NFS4_MAXLABELLEN has defined for sec label max length, use it directly. Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2015-03-28 23:46:09 +08:00
ret += NFS4_MAXLABELLEN + 12;
nfsd4: better estimate of getattr response size We plan to use this estimate to decide whether or not to allow zero-copy reads. Currently we're assuming all getattr's are a page, which can be both too small (ACLs e.g. may be arbitrarily long) and too large (after an upcoming read patch this will unnecessarily prevent zero copy reads in any read compound also containing a getattr). Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-24 00:01:48 +08:00
bmap2 &= ~FATTR4_WORD2_SECURITY_LABEL;
}
/*
* Largest of remaining attributes are 16 bytes (e.g.,
* supported_attributes)
*/
ret += 16 * (hweight32(bmap0) + hweight32(bmap1) + hweight32(bmap2));
/* bitmask, length */
ret += 20;
return ret;
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_getfh_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
NFSD: Get response size before operation for all RPCs NFSD usess PAGE_SIZE as the reply size estimate for RPCs which don't support op_rsize_bop(), A PAGE_SIZE (4096) is larger than many real response sizes, eg, access (op_encode_hdr_size + 2), seek (op_encode_hdr_size + 3). This patch just adds op_rsize_bop() for all RPCs getting response size. An overestimate is generally safe but the tighter estimates are probably better. Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-02-03 22:36:00 +08:00
{
return (op_encode_hdr_size + 1) * sizeof(__be32) + NFS4_FHSIZE;
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_link_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
{
return (op_encode_hdr_size + op_encode_change_info_maxsz)
* sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_lock_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
{
return (op_encode_hdr_size + op_encode_lock_denied_maxsz)
* sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_open_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
{
return (op_encode_hdr_size + op_encode_stateid_maxsz
+ op_encode_change_info_maxsz + 1
+ nfs4_fattr_bitmap_maxsz
+ op_encode_delegation_maxsz) * sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_read_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
{
NFSD: Cap rsize_bop result based on send buffer size Since before the git era, NFSD has conserved the number of pages held by each nfsd thread by combining the RPC receive and send buffers into a single array of pages. This works because there are no cases where an operation needs a large RPC Call message and a large RPC Reply at the same time. Once an RPC Call has been received, svc_process() updates svc_rqst::rq_res to describe the part of rq_pages that can be used for constructing the Reply. This means that the send buffer (rq_res) shrinks when the received RPC record containing the RPC Call is large. Add an NFSv4 helper that computes the size of the send buffer. It replaces svc_max_payload() in spots where svc_max_payload() returns a value that might be larger than the remaining send buffer space. Callers who need to know the transport's actual maximum payload size will continue to use svc_max_payload(). Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-02 03:29:55 +08:00
u32 rlen = min(op->u.read.rd_length, nfsd4_max_payload(rqstp));
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
nfsd4: read size estimate should include padding Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-05-17 02:38:20 +08:00
return (op_encode_hdr_size + 2 + XDR_QUADLEN(rlen)) * sizeof(__be32);
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_read_plus_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
NFSD: Add READ_PLUS data support This patch adds READ_PLUS support for returning a single NFS4_CONTENT_DATA segment to the client. This is basically the same as the READ operation, only with the extra information about data segments. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2020-09-29 01:08:58 +08:00
{
NFSD: Cap rsize_bop result based on send buffer size Since before the git era, NFSD has conserved the number of pages held by each nfsd thread by combining the RPC receive and send buffers into a single array of pages. This works because there are no cases where an operation needs a large RPC Call message and a large RPC Reply at the same time. Once an RPC Call has been received, svc_process() updates svc_rqst::rq_res to describe the part of rq_pages that can be used for constructing the Reply. This means that the send buffer (rq_res) shrinks when the received RPC record containing the RPC Call is large. Add an NFSv4 helper that computes the size of the send buffer. It replaces svc_max_payload() in spots where svc_max_payload() returns a value that might be larger than the remaining send buffer space. Callers who need to know the transport's actual maximum payload size will continue to use svc_max_payload(). Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-02 03:29:55 +08:00
u32 rlen = min(op->u.read.rd_length, nfsd4_max_payload(rqstp));
NFSD: Add READ_PLUS data support This patch adds READ_PLUS support for returning a single NFS4_CONTENT_DATA segment to the client. This is basically the same as the READ operation, only with the extra information about data segments. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2020-09-29 01:08:58 +08:00
/*
* If we detect that the file changed during hole encoding, then we
* recover by encoding the remaining reply as data. This means we need
* to set aside enough room to encode two data segments.
*/
u32 seg_len = 2 * (1 + 2 + 1);
return (op_encode_hdr_size + 2 + seg_len + XDR_QUADLEN(rlen)) * sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_readdir_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
{
NFSD: Cap rsize_bop result based on send buffer size Since before the git era, NFSD has conserved the number of pages held by each nfsd thread by combining the RPC receive and send buffers into a single array of pages. This works because there are no cases where an operation needs a large RPC Call message and a large RPC Reply at the same time. Once an RPC Call has been received, svc_process() updates svc_rqst::rq_res to describe the part of rq_pages that can be used for constructing the Reply. This means that the send buffer (rq_res) shrinks when the received RPC record containing the RPC Call is large. Add an NFSv4 helper that computes the size of the send buffer. It replaces svc_max_payload() in spots where svc_max_payload() returns a value that might be larger than the remaining send buffer space. Callers who need to know the transport's actual maximum payload size will continue to use svc_max_payload(). Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-02 03:29:55 +08:00
u32 rlen = min(op->u.readdir.rd_maxcount, nfsd4_max_payload(rqstp));
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
nfsd4: allow large readdirs Currently we limit readdir results to a single page. This can result in a performance regression compared to NFSv3 when reading large directories. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-01-21 05:37:11 +08:00
return (op_encode_hdr_size + op_encode_verifier_maxsz +
XDR_QUADLEN(rlen)) * sizeof(__be32);
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_readlink_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
NFSD: Get response size before operation for all RPCs NFSD usess PAGE_SIZE as the reply size estimate for RPCs which don't support op_rsize_bop(), A PAGE_SIZE (4096) is larger than many real response sizes, eg, access (op_encode_hdr_size + 2), seek (op_encode_hdr_size + 3). This patch just adds op_rsize_bop() for all RPCs getting response size. An overestimate is generally safe but the tighter estimates are probably better. Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-02-03 22:36:00 +08:00
{
return (op_encode_hdr_size + 1) * sizeof(__be32) + PAGE_SIZE;
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_remove_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
{
return (op_encode_hdr_size + op_encode_change_info_maxsz)
* sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_rename_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
{
return (op_encode_hdr_size + op_encode_change_info_maxsz
+ op_encode_change_info_maxsz) * sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_sequence_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd4: estimate sequence response size Otherwise a following patch would turn off all 4.1 zero-copy reads. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-24 00:34:22 +08:00
{
nfsd4: fix response size estimation for OP_SEQUENCE We added this new estimator function but forgot to hook it up. The effect is that NFSv4.1 (and greater) won't do zero-copy reads. The estimate was also wrong by 8 bytes. Fixes: ccae70a9ee41 "nfsd4: estimate sequence response size" Cc: stable@vger.kernel.org Reported-by: Chuck Lever <chucklever@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-08-22 03:04:31 +08:00
return (op_encode_hdr_size
+ XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + 5) * sizeof(__be32);
nfsd4: estimate sequence response size Otherwise a following patch would turn off all 4.1 zero-copy reads. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-24 00:34:22 +08:00
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_test_stateid_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
NFSD: Get response size before operation for all RPCs NFSD usess PAGE_SIZE as the reply size estimate for RPCs which don't support op_rsize_bop(), A PAGE_SIZE (4096) is larger than many real response sizes, eg, access (op_encode_hdr_size + 2), seek (op_encode_hdr_size + 3). This patch just adds op_rsize_bop() for all RPCs getting response size. An overestimate is generally safe but the tighter estimates are probably better. Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-02-03 22:36:00 +08:00
{
return (op_encode_hdr_size + 1 + op->u.test_stateid.ts_num_ids)
* sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_setattr_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
{
return (op_encode_hdr_size + nfs4_fattr_bitmap_maxsz) * sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_secinfo_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
NFSD: Get response size before operation for all RPCs NFSD usess PAGE_SIZE as the reply size estimate for RPCs which don't support op_rsize_bop(), A PAGE_SIZE (4096) is larger than many real response sizes, eg, access (op_encode_hdr_size + 2), seek (op_encode_hdr_size + 3). This patch just adds op_rsize_bop() for all RPCs getting response size. An overestimate is generally safe but the tighter estimates are probably better. Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-02-03 22:36:00 +08:00
{
return (op_encode_hdr_size + RPC_AUTH_MAXFLAVOR *
(4 + XDR_QUADLEN(GSS_OID_MAX_LEN))) * sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_setclientid_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
{
nfsd4: fix setclientid encode size Cc: stable@vger.kernel.org Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-11 02:17:55 +08:00
return (op_encode_hdr_size + 2 + XDR_QUADLEN(NFS4_VERIFIER_SIZE)) *
sizeof(__be32);
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_write_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
{
nfsd4: fix write reply size estimate The write reply also includes count and stable_how. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-05-17 09:24:56 +08:00
return (op_encode_hdr_size + 2 + op_encode_verifier_maxsz) * sizeof(__be32);
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_exchange_id_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
{
return (op_encode_hdr_size + 2 + 1 + /* eir_clientid, eir_sequenceid */\
nfsd: calculate the missing length of bitmap in EXCHANGE_ID commit 58cd57bfd9db3bc213bf9d6a10920f82095f0114 "nfsd: Fix SP4_MACH_CRED negotiation in EXCHANGE_ID" miss calculating the length of bitmap for spo_must_enforce and spo_must_allow. Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2013-12-10 15:24:36 +08:00
1 + 1 + /* eir_flags, spr_how */\
4 + /* spo_must_enforce & _allow with bitmap */\
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
2 + /*eir_server_owner.so_minor_id */\
/* eir_server_owner.so_major_id<> */\
XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + 1 +\
/* eir_server_scope<> */\
XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + 1 +\
1 + /* eir_server_impl_id array length */\
0 /* ignored eir_server_impl_id contents */) * sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_bind_conn_to_session_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
{
return (op_encode_hdr_size + \
XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + /* bctsr_sessid */\
2 /* bctsr_dir, use_conn_in_rdma_mode */) * sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_create_session_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
{
return (op_encode_hdr_size + \
XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + /* sessionid */\
2 + /* csr_sequence, csr_flags */\
op_encode_channel_attrs_maxsz + \
op_encode_channel_attrs_maxsz) * sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_copy_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
NFSD: Implement the COPY call I only implemented the sync version of this call, since it's the easiest. I can simply call vfs_copy_range() and have the vfs do the right thing for the filesystem being exported. Signed-off-by: Anna Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-09-08 03:57:30 +08:00
{
return (op_encode_hdr_size +
1 /* wr_callback */ +
op_encode_stateid_maxsz /* wr_callback */ +
2 /* wr_count */ +
1 /* wr_committed */ +
op_encode_verifier_maxsz +
1 /* cr_consecutive */ +
1 /* cr_synchronous */) * sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_offload_status_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
NFSD OFFLOAD_STATUS xdr Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:18 +08:00
{
return (op_encode_hdr_size +
2 /* osr_count */ +
1 /* osr_complete<1> optional 0 for now */) * sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_copy_notify_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
NFSD COPY_NOTIFY xdr Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-08-08 23:14:59 +08:00
{
return (op_encode_hdr_size +
3 /* cnr_lease_time */ +
1 /* We support one cnr_source_server */ +
1 /* cnr_stateid seq */ +
op_encode_stateid_maxsz /* cnr_stateid */ +
1 /* num cnr_source_server*/ +
1 /* nl4_type */ +
1 /* nl4 size */ +
XDR_QUADLEN(NFS4_OPAQUE_LIMIT) /*nl4_loc + nl4_loc_sz */)
* sizeof(__be32);
}
nfsd: trivial GET_DIR_DELEGATION support This adds basic infrastructure for handing GET_DIR_DELEGATION calls from clients, including the decoders and encoders. For now, it always just returns NFS4_OK + GDD4_UNAVAIL. Eventually clients may start sending this operation, and it's better if we can return GDD4_UNAVAIL instead of having to abort the whole compound. Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-03-23 21:25:54 +08:00
static u32 nfsd4_get_dir_delegation_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
{
return (op_encode_hdr_size +
1 /* gddr_status */ +
op_encode_verifier_maxsz +
op_encode_stateid_maxsz +
2 /* gddr_notification */ +
2 /* gddr_child_attributes */ +
2 /* gddr_dir_attributes */);
}
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
#ifdef CONFIG_NFSD_PNFS
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_getdeviceinfo_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
NFSD: Get response size before operation for all RPCs NFSD usess PAGE_SIZE as the reply size estimate for RPCs which don't support op_rsize_bop(), A PAGE_SIZE (4096) is larger than many real response sizes, eg, access (op_encode_hdr_size + 2), seek (op_encode_hdr_size + 3). This patch just adds op_rsize_bop() for all RPCs getting response size. An overestimate is generally safe but the tighter estimates are probably better. Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-02-03 22:36:00 +08:00
{
NFSD: Cap rsize_bop result based on send buffer size Since before the git era, NFSD has conserved the number of pages held by each nfsd thread by combining the RPC receive and send buffers into a single array of pages. This works because there are no cases where an operation needs a large RPC Call message and a large RPC Reply at the same time. Once an RPC Call has been received, svc_process() updates svc_rqst::rq_res to describe the part of rq_pages that can be used for constructing the Reply. This means that the send buffer (rq_res) shrinks when the received RPC record containing the RPC Call is large. Add an NFSv4 helper that computes the size of the send buffer. It replaces svc_max_payload() in spots where svc_max_payload() returns a value that might be larger than the remaining send buffer space. Callers who need to know the transport's actual maximum payload size will continue to use svc_max_payload(). Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-02 03:29:55 +08:00
u32 rlen = min(op->u.getdeviceinfo.gd_maxcount, nfsd4_max_payload(rqstp));
NFSD: Get response size before operation for all RPCs NFSD usess PAGE_SIZE as the reply size estimate for RPCs which don't support op_rsize_bop(), A PAGE_SIZE (4096) is larger than many real response sizes, eg, access (op_encode_hdr_size + 2), seek (op_encode_hdr_size + 3). This patch just adds op_rsize_bop() for all RPCs getting response size. An overestimate is generally safe but the tighter estimates are probably better. Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-02-03 22:36:00 +08:00
return (op_encode_hdr_size +
1 /* gd_layout_type*/ +
XDR_QUADLEN(rlen) +
2 /* gd_notify_types */) * sizeof(__be32);
}
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
/*
* At this stage we don't really know what layout driver will handle the request,
* so we need to define an arbitrary upper bound here.
*/
#define MAX_LAYOUT_SIZE 128
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_layoutget_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
{
return (op_encode_hdr_size +
1 /* logr_return_on_close */ +
op_encode_stateid_maxsz +
1 /* nr of layouts */ +
MAX_LAYOUT_SIZE) * sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_layoutcommit_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
{
return (op_encode_hdr_size +
1 /* locr_newsize */ +
2 /* ns_size */) * sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_layoutreturn_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
{
return (op_encode_hdr_size +
1 /* lrs_stateid */ +
op_encode_stateid_maxsz) * sizeof(__be32);
}
#endif /* CONFIG_NFSD_PNFS */
NFSD: Get response size before operation for all RPCs NFSD usess PAGE_SIZE as the reply size estimate for RPCs which don't support op_rsize_bop(), A PAGE_SIZE (4096) is larger than many real response sizes, eg, access (op_encode_hdr_size + 2), seek (op_encode_hdr_size + 3). This patch just adds op_rsize_bop() for all RPCs getting response size. An overestimate is generally safe but the tighter estimates are probably better. Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-02-03 22:36:00 +08:00
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_seek_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
NFSD: Get response size before operation for all RPCs NFSD usess PAGE_SIZE as the reply size estimate for RPCs which don't support op_rsize_bop(), A PAGE_SIZE (4096) is larger than many real response sizes, eg, access (op_encode_hdr_size + 2), seek (op_encode_hdr_size + 3). This patch just adds op_rsize_bop() for all RPCs getting response size. An overestimate is generally safe but the tighter estimates are probably better. Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-02-03 22:36:00 +08:00
{
return (op_encode_hdr_size + 3) * sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_getxattr_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd: implement the xattr functions and en/decode logic Implement the main entry points for the *XATTR operations. Add functions to calculate the reply size for the user extended attribute operations, and implement the XDR encode / decode logic for these operations. Add the user extended attributes operations to nfsd4_ops. Signed-off-by: Frank van der Linden <fllinden@amazon.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-06-24 06:39:26 +08:00
{
NFSD: Cap rsize_bop result based on send buffer size Since before the git era, NFSD has conserved the number of pages held by each nfsd thread by combining the RPC receive and send buffers into a single array of pages. This works because there are no cases where an operation needs a large RPC Call message and a large RPC Reply at the same time. Once an RPC Call has been received, svc_process() updates svc_rqst::rq_res to describe the part of rq_pages that can be used for constructing the Reply. This means that the send buffer (rq_res) shrinks when the received RPC record containing the RPC Call is large. Add an NFSv4 helper that computes the size of the send buffer. It replaces svc_max_payload() in spots where svc_max_payload() returns a value that might be larger than the remaining send buffer space. Callers who need to know the transport's actual maximum payload size will continue to use svc_max_payload(). Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-02 03:29:55 +08:00
u32 rlen = min_t(u32, XATTR_SIZE_MAX, nfsd4_max_payload(rqstp));
nfsd: implement the xattr functions and en/decode logic Implement the main entry points for the *XATTR operations. Add functions to calculate the reply size for the user extended attribute operations, and implement the XDR encode / decode logic for these operations. Add the user extended attributes operations to nfsd4_ops. Signed-off-by: Frank van der Linden <fllinden@amazon.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-06-24 06:39:26 +08:00
return (op_encode_hdr_size + 1 + XDR_QUADLEN(rlen)) * sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_setxattr_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd: implement the xattr functions and en/decode logic Implement the main entry points for the *XATTR operations. Add functions to calculate the reply size for the user extended attribute operations, and implement the XDR encode / decode logic for these operations. Add the user extended attributes operations to nfsd4_ops. Signed-off-by: Frank van der Linden <fllinden@amazon.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-06-24 06:39:26 +08:00
{
return (op_encode_hdr_size + op_encode_change_info_maxsz)
* sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_listxattrs_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd: implement the xattr functions and en/decode logic Implement the main entry points for the *XATTR operations. Add functions to calculate the reply size for the user extended attribute operations, and implement the XDR encode / decode logic for these operations. Add the user extended attributes operations to nfsd4_ops. Signed-off-by: Frank van der Linden <fllinden@amazon.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-06-24 06:39:26 +08:00
{
NFSD: Cap rsize_bop result based on send buffer size Since before the git era, NFSD has conserved the number of pages held by each nfsd thread by combining the RPC receive and send buffers into a single array of pages. This works because there are no cases where an operation needs a large RPC Call message and a large RPC Reply at the same time. Once an RPC Call has been received, svc_process() updates svc_rqst::rq_res to describe the part of rq_pages that can be used for constructing the Reply. This means that the send buffer (rq_res) shrinks when the received RPC record containing the RPC Call is large. Add an NFSv4 helper that computes the size of the send buffer. It replaces svc_max_payload() in spots where svc_max_payload() returns a value that might be larger than the remaining send buffer space. Callers who need to know the transport's actual maximum payload size will continue to use svc_max_payload(). Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-02 03:29:55 +08:00
u32 rlen = min(op->u.listxattrs.lsxa_maxcount, nfsd4_max_payload(rqstp));
nfsd: implement the xattr functions and en/decode logic Implement the main entry points for the *XATTR operations. Add functions to calculate the reply size for the user extended attribute operations, and implement the XDR encode / decode logic for these operations. Add the user extended attributes operations to nfsd4_ops. Signed-off-by: Frank van der Linden <fllinden@amazon.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-06-24 06:39:26 +08:00
return (op_encode_hdr_size + 4 + XDR_QUADLEN(rlen)) * sizeof(__be32);
}
NFSD: Remove "inline" directives on op_rsize_bop helpers These helpers are always invoked indirectly, so the compiler can't inline these anyway. While we're updating the synopses of these helpers, defensively convert their parameters to const pointers. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:23:25 +08:00
static u32 nfsd4_removexattr_rsize(const struct svc_rqst *rqstp,
const struct nfsd4_op *op)
nfsd: implement the xattr functions and en/decode logic Implement the main entry points for the *XATTR operations. Add functions to calculate the reply size for the user extended attribute operations, and implement the XDR encode / decode logic for these operations. Add the user extended attributes operations to nfsd4_ops. Signed-off-by: Frank van der Linden <fllinden@amazon.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-06-24 06:39:26 +08:00
{
return (op_encode_hdr_size + op_encode_change_info_maxsz)
* sizeof(__be32);
}
nfsd4: const-ify nfsd4_ops nfsd4_ops contains function pointers, and marking it as constant avoids it being able to be used as an attach vector for code injections. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 05:46:47 +08:00
static const struct nfsd4_operation nfsd4_ops[] = {
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
[OP_ACCESS] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_access,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_ACCESS",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_access_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_CLOSE] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_close,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_flags = OP_MODIFIES_SOMETHING,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_CLOSE",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_status_stateid_rsize,
nfsd4: properly type op_get_currentstateid callbacks Pass union nfsd4_op_u to the op_set_currentstateid callbacks instead of using unsafe function pointer casts. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:37:33 +08:00
.op_get_currentstateid = nfsd4_get_closestateid,
nfsd4: properly type op_set_currentstateid callbacks Given the args union in struct nfsd4_op a name, and pass it to the op_set_currentstateid callbacks instead of using unsafe function pointer casts. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:03:15 +08:00
.op_set_currentstateid = nfsd4_set_closestateid,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_COMMIT] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_commit,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_flags = OP_MODIFIES_SOMETHING,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_COMMIT",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_commit_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_CREATE] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_create,
nfsd41: mark LOOKUP, LOOKUPP and CREATE to invalidate current stateid Signed-off-by: Tigran Mkrtchyan <kofemann@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-02-14 05:55:29 +08:00
.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME | OP_CLEAR_STATEID,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_CREATE",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_create_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_DELEGRETURN] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_delegreturn,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_flags = OP_MODIFIES_SOMETHING,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_DELEGRETURN",
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
nfsd4: properly type op_get_currentstateid callbacks Pass union nfsd4_op_u to the op_set_currentstateid callbacks instead of using unsafe function pointer casts. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:37:33 +08:00
.op_get_currentstateid = nfsd4_get_delegreturnstateid,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_GETATTR] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_getattr,
[PATCH] knfsd: nfsd4: simplify migration op check I'm not too fond of these big if conditions. Replace them by checks of a flag in the operation descriptor. To my eye this makes the code a bit more self-documenting, and makes the complicated part of the code (proc_compound) a little more compact. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:39 +08:00
.op_flags = ALLOWED_ON_ABSENT_FS,
nfsd4: better estimate of getattr response size We plan to use this estimate to decide whether or not to allow zero-copy reads. Currently we're assuming all getattr's are a page, which can be both too small (ACLs e.g. may be arbitrarily long) and too large (after an upcoming read patch this will unnecessarily prevent zero copy reads in any read compound also containing a getattr). Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-24 00:01:48 +08:00
.op_rsize_bop = nfsd4_getattr_rsize,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_GETATTR",
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_GETFH] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_getfh,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_GETFH",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_getfh_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_LINK] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_link,
nfsd4: make op_cacheresult another flag I'm not sure why I used a new field for this originally. Also, the differences between some of these flags are a little subtle; add some comments to explain. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-09-20 20:49:51 +08:00
.op_flags = ALLOWED_ON_ABSENT_FS | OP_MODIFIES_SOMETHING
| OP_CACHEME,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_LINK",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_link_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_LOCK] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_lock,
NFSD: Remove a layering violation when encoding lock_denied An XDR encoder is responsible for marshaling results, not releasing memory that was allocated by the upper layer. We have .op_release for that purpose. Move the release of the ld_owner.data string to op_release functions for LOCK and LOCKT. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-10-13 01:46:39 +08:00
.op_release = nfsd4_lock_release,
nfsd4: skip encoder in trivial error cases Most encoders do nothing in the error case. But they can still screw things up in that case: most errors happen very early in rpc processing, possibly before argument fields are filled in and bounds-tested, so encoders that do anything other than immediately bail on error can easily crash in odd error cases. So just handle errors centrally most of the time to remove the chance of error. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-05-06 22:49:21 +08:00
.op_flags = OP_MODIFIES_SOMETHING |
OP_NONTRIVIAL_ERROR_ENCODE,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_LOCK",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_lock_rsize,
nfsd4: properly type op_set_currentstateid callbacks Given the args union in struct nfsd4_op a name, and pass it to the op_set_currentstateid callbacks instead of using unsafe function pointer casts. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:03:15 +08:00
.op_set_currentstateid = nfsd4_set_lockstateid,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_LOCKT] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_lockt,
NFSD: Remove a layering violation when encoding lock_denied An XDR encoder is responsible for marshaling results, not releasing memory that was allocated by the upper layer. We have .op_release for that purpose. Move the release of the ld_owner.data string to op_release functions for LOCK and LOCKT. Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-10-13 01:46:39 +08:00
.op_release = nfsd4_lockt_release,
nfsd4: skip encoder in trivial error cases Most encoders do nothing in the error case. But they can still screw things up in that case: most errors happen very early in rpc processing, possibly before argument fields are filled in and bounds-tested, so encoders that do anything other than immediately bail on error can easily crash in odd error cases. So just handle errors centrally most of the time to remove the chance of error. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-05-06 22:49:21 +08:00
.op_flags = OP_NONTRIVIAL_ERROR_ENCODE,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_LOCKT",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_lock_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_LOCKU] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_locku,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_flags = OP_MODIFIES_SOMETHING,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_LOCKU",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_status_stateid_rsize,
nfsd4: properly type op_get_currentstateid callbacks Pass union nfsd4_op_u to the op_set_currentstateid callbacks instead of using unsafe function pointer casts. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:37:33 +08:00
.op_get_currentstateid = nfsd4_get_lockustateid,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_LOOKUP] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_lookup,
nfsd41: mark LOOKUP, LOOKUPP and CREATE to invalidate current stateid Signed-off-by: Tigran Mkrtchyan <kofemann@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-02-14 05:55:29 +08:00
.op_flags = OP_HANDLES_WRONGSEC | OP_CLEAR_STATEID,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_LOOKUP",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_LOOKUPP] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_lookupp,
nfsd41: mark LOOKUP, LOOKUPP and CREATE to invalidate current stateid Signed-off-by: Tigran Mkrtchyan <kofemann@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-02-14 05:55:29 +08:00
.op_flags = OP_HANDLES_WRONGSEC | OP_CLEAR_STATEID,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_LOOKUPP",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_NVERIFY] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_nverify,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_NVERIFY",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_OPEN] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_open,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_flags = OP_HANDLES_WRONGSEC | OP_MODIFIES_SOMETHING,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_OPEN",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_open_rsize,
nfsd4: properly type op_set_currentstateid callbacks Given the args union in struct nfsd4_op a name, and pass it to the op_set_currentstateid callbacks instead of using unsafe function pointer casts. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:03:15 +08:00
.op_set_currentstateid = nfsd4_set_openstateid,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_OPEN_CONFIRM] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_open_confirm,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_flags = OP_MODIFIES_SOMETHING,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_OPEN_CONFIRM",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_status_stateid_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_OPEN_DOWNGRADE] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_open_downgrade,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_flags = OP_MODIFIES_SOMETHING,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_OPEN_DOWNGRADE",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_status_stateid_rsize,
nfsd4: properly type op_get_currentstateid callbacks Pass union nfsd4_op_u to the op_set_currentstateid callbacks instead of using unsafe function pointer casts. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:37:33 +08:00
.op_get_currentstateid = nfsd4_get_opendowngradestateid,
nfsd4: properly type op_set_currentstateid callbacks Given the args union in struct nfsd4_op a name, and pass it to the op_set_currentstateid callbacks instead of using unsafe function pointer casts. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:03:15 +08:00
.op_set_currentstateid = nfsd4_set_opendowngradestateid,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_PUTFH] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_putfh,
nfsd4: fix wrongsec handling for PUTFH + op cases When PUTFH is followed by an operation that uses the filehandle, and when the current client is using a security flavor that is inconsistent with the given filehandle, we have a choice: we can return WRONGSEC either when the current filehandle is set using the PUTFH, or when the filehandle is first used by the following operation. Follow the recommendations of RFC 5661 in making this choice. (Our current behavior prevented the client from doing security negotiation by returning WRONGSEC on PUTFH+SECINFO_NO_NAME.) Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-04-09 05:00:50 +08:00
.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
nfsd4: READ, READDIR, etc., are idempotent OP_MODIFIES_SOMETHING flags operations that we should be careful not to initiate without being sure we have the buffer space to encode a reply. None of these ops fall into that category. We could probably remove a few more, but this isn't a very important problem at least for ops whose reply size is easy to estimate. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-08 00:43:58 +08:00
| OP_IS_PUTFH_LIKE | OP_CLEAR_STATEID,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_PUTFH",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[PATCH] knfsd: nfsd4: simplify migration op check I'm not too fond of these big if conditions. Replace them by checks of a flag in the operation descriptor. To my eye this makes the code a bit more self-documenting, and makes the complicated part of the code (proc_compound) a little more compact. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:39 +08:00
[OP_PUTPUBFH] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_putrootfh,
nfsd4: fix wrongsec handling for PUTFH + op cases When PUTFH is followed by an operation that uses the filehandle, and when the current client is using a security flavor that is inconsistent with the given filehandle, we have a choice: we can return WRONGSEC either when the current filehandle is set using the PUTFH, or when the filehandle is first used by the following operation. Follow the recommendations of RFC 5661 in making this choice. (Our current behavior prevented the client from doing security negotiation by returning WRONGSEC on PUTFH+SECINFO_NO_NAME.) Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-04-09 05:00:50 +08:00
.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
nfsd4: READ, READDIR, etc., are idempotent OP_MODIFIES_SOMETHING flags operations that we should be careful not to initiate without being sure we have the buffer space to encode a reply. None of these ops fall into that category. We could probably remove a few more, but this isn't a very important problem at least for ops whose reply size is easy to estimate. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-08 00:43:58 +08:00
| OP_IS_PUTFH_LIKE | OP_CLEAR_STATEID,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_PUTPUBFH",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
[PATCH] knfsd: nfsd4: simplify migration op check I'm not too fond of these big if conditions. Replace them by checks of a flag in the operation descriptor. To my eye this makes the code a bit more self-documenting, and makes the complicated part of the code (proc_compound) a little more compact. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:39 +08:00
},
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
[OP_PUTROOTFH] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_putrootfh,
nfsd4: fix wrongsec handling for PUTFH + op cases When PUTFH is followed by an operation that uses the filehandle, and when the current client is using a security flavor that is inconsistent with the given filehandle, we have a choice: we can return WRONGSEC either when the current filehandle is set using the PUTFH, or when the filehandle is first used by the following operation. Follow the recommendations of RFC 5661 in making this choice. (Our current behavior prevented the client from doing security negotiation by returning WRONGSEC on PUTFH+SECINFO_NO_NAME.) Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-04-09 05:00:50 +08:00
.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
nfsd4: READ, READDIR, etc., are idempotent OP_MODIFIES_SOMETHING flags operations that we should be careful not to initiate without being sure we have the buffer space to encode a reply. None of these ops fall into that category. We could probably remove a few more, but this isn't a very important problem at least for ops whose reply size is easy to estimate. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-08 00:43:58 +08:00
| OP_IS_PUTFH_LIKE | OP_CLEAR_STATEID,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_PUTROOTFH",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_READ] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_read,
nfsd4: define ->op_release for compound ops Run a separate ->op_release function if necessary instead of depending on the xdr encoder to do this. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-05-06 05:09:37 +08:00
.op_release = nfsd4_read_release,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_READ",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_read_rsize,
nfsd4: properly type op_get_currentstateid callbacks Pass union nfsd4_op_u to the op_set_currentstateid callbacks instead of using unsafe function pointer casts. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:37:33 +08:00
.op_get_currentstateid = nfsd4_get_readstateid,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_READDIR] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_readdir,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_READDIR",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_readdir_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_READLINK] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_readlink,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_READLINK",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_readlink_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_REMOVE] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_remove,
nfsd4: make op_cacheresult another flag I'm not sure why I used a new field for this originally. Also, the differences between some of these flags are a little subtle; add some comments to explain. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-09-20 20:49:51 +08:00
.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_REMOVE",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_remove_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_RENAME] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_rename,
nfsd4: make op_cacheresult another flag I'm not sure why I used a new field for this originally. Also, the differences between some of these flags are a little subtle; add some comments to explain. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-09-20 20:49:51 +08:00
.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_name = "OP_RENAME",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_rename_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_RENEW] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_renew,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
| OP_MODIFIES_SOMETHING,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_RENEW",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_RESTOREFH] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_restorefh,
nfsd4: fix wrongsec handling for PUTFH + op cases When PUTFH is followed by an operation that uses the filehandle, and when the current client is using a security flavor that is inconsistent with the given filehandle, we have a choice: we can return WRONGSEC either when the current filehandle is set using the PUTFH, or when the filehandle is first used by the following operation. Follow the recommendations of RFC 5661 in making this choice. (Our current behavior prevented the client from doing security negotiation by returning WRONGSEC on PUTFH+SECINFO_NO_NAME.) Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-04-09 05:00:50 +08:00
.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
| OP_IS_PUTFH_LIKE | OP_MODIFIES_SOMETHING,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_RESTOREFH",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_SAVEFH] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_savefh,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_flags = OP_HANDLES_WRONGSEC | OP_MODIFIES_SOMETHING,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_SAVEFH",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
knfsd: nfsd4: implement secinfo Implement the secinfo operation. (Thanks to Usha Ketineni wrote an earlier version of this support.) Cc: Usha Ketineni <uketinen@us.ibm.com> Signed-off-by: Andy Adamson <andros@citi.umich.edu> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-17 19:04:51 +08:00
[OP_SECINFO] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_secinfo,
nfsd4: define ->op_release for compound ops Run a separate ->op_release function if necessary instead of depending on the xdr encoder to do this. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-05-06 05:09:37 +08:00
.op_release = nfsd4_secinfo_release,
nfsd4: fix wrongsec handling for PUTFH + op cases When PUTFH is followed by an operation that uses the filehandle, and when the current client is using a security flavor that is inconsistent with the given filehandle, we have a choice: we can return WRONGSEC either when the current filehandle is set using the PUTFH, or when the filehandle is first used by the following operation. Follow the recommendations of RFC 5661 in making this choice. (Our current behavior prevented the client from doing security negotiation by returning WRONGSEC on PUTFH+SECINFO_NO_NAME.) Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-04-09 05:00:50 +08:00
.op_flags = OP_HANDLES_WRONGSEC,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_SECINFO",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_secinfo_rsize,
knfsd: nfsd4: implement secinfo Implement the secinfo operation. (Thanks to Usha Ketineni wrote an earlier version of this support.) Cc: Usha Ketineni <uketinen@us.ibm.com> Signed-off-by: Andy Adamson <andros@citi.umich.edu> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-17 19:04:51 +08:00
},
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
[OP_SETATTR] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_setattr,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_SETATTR",
nfsd4: skip encoder in trivial error cases Most encoders do nothing in the error case. But they can still screw things up in that case: most errors happen very early in rpc processing, possibly before argument fields are filled in and bounds-tested, so encoders that do anything other than immediately bail on error can easily crash in odd error cases. So just handle errors centrally most of the time to remove the chance of error. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-05-06 22:49:21 +08:00
.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME
| OP_NONTRIVIAL_ERROR_ENCODE,
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_setattr_rsize,
nfsd4: properly type op_get_currentstateid callbacks Pass union nfsd4_op_u to the op_set_currentstateid callbacks instead of using unsafe function pointer casts. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:37:33 +08:00
.op_get_currentstateid = nfsd4_get_setattrstateid,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_SETCLIENTID] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_setclientid,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
nfsd4: skip encoder in trivial error cases Most encoders do nothing in the error case. But they can still screw things up in that case: most errors happen very early in rpc processing, possibly before argument fields are filled in and bounds-tested, so encoders that do anything other than immediately bail on error can easily crash in odd error cases. So just handle errors centrally most of the time to remove the chance of error. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-05-06 22:49:21 +08:00
| OP_MODIFIES_SOMETHING | OP_CACHEME
| OP_NONTRIVIAL_ERROR_ENCODE,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_SETCLIENTID",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_setclientid_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_SETCLIENTID_CONFIRM] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_setclientid_confirm,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
nfsd4: make op_cacheresult another flag I'm not sure why I used a new field for this originally. Also, the differences between some of these flags are a little subtle; add some comments to explain. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-09-20 20:49:51 +08:00
| OP_MODIFIES_SOMETHING | OP_CACHEME,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_SETCLIENTID_CONFIRM",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_VERIFY] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_verify,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_VERIFY",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_WRITE] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_write,
nfsd4: make op_cacheresult another flag I'm not sure why I used a new field for this originally. Also, the differences between some of these flags are a little subtle; add some comments to explain. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-09-20 20:49:51 +08:00
.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_WRITE",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_write_rsize,
nfsd4: properly type op_get_currentstateid callbacks Pass union nfsd4_op_u to the op_set_currentstateid callbacks instead of using unsafe function pointer casts. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:37:33 +08:00
.op_get_currentstateid = nfsd4_get_writestateid,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
[OP_RELEASE_LOCKOWNER] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_release_lockowner,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
| OP_MODIFIES_SOMETHING,
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
.op_name = "OP_RELEASE_LOCKOWNER",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
},
nfsd41: proc stubs Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:27:58 +08:00
/* NFSv4.1 operations */
[OP_EXCHANGE_ID] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_exchange_id,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
| OP_MODIFIES_SOMETHING,
nfsd41: proc stubs Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:27:58 +08:00
.op_name = "OP_EXCHANGE_ID",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_exchange_id_rsize,
nfsd41: proc stubs Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:27:58 +08:00
},
nfsd4: implement backchannel_ctl operation This operation is mandatory for servers to implement. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-11-02 06:09:48 +08:00
[OP_BACKCHANNEL_CTL] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_backchannel_ctl,
nfsd4: implement backchannel_ctl operation This operation is mandatory for servers to implement. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-11-02 06:09:48 +08:00
.op_flags = ALLOWED_WITHOUT_FH | OP_MODIFIES_SOMETHING,
.op_name = "OP_BACKCHANNEL_CTL",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
nfsd4: implement backchannel_ctl operation This operation is mandatory for servers to implement. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-11-02 06:09:48 +08:00
},
nfsd4: support BIND_CONN_TO_SESSION Basic xdr and processing for BIND_CONN_TO_SESSION. This adds a connection to the list of connections associated with a session. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2010-10-05 11:12:59 +08:00
[OP_BIND_CONN_TO_SESSION] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_bind_conn_to_session,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
| OP_MODIFIES_SOMETHING,
nfsd4: support BIND_CONN_TO_SESSION Basic xdr and processing for BIND_CONN_TO_SESSION. This adds a connection to the list of connections associated with a session. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2010-10-05 11:12:59 +08:00
.op_name = "OP_BIND_CONN_TO_SESSION",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_bind_conn_to_session_rsize,
nfsd4: support BIND_CONN_TO_SESSION Basic xdr and processing for BIND_CONN_TO_SESSION. This adds a connection to the list of connections associated with a session. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2010-10-05 11:12:59 +08:00
},
nfsd41: proc stubs Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:27:58 +08:00
[OP_CREATE_SESSION] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_create_session,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
| OP_MODIFIES_SOMETHING,
nfsd41: proc stubs Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:27:58 +08:00
.op_name = "OP_CREATE_SESSION",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_create_session_rsize,
nfsd41: proc stubs Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:27:58 +08:00
},
[OP_DESTROY_SESSION] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_destroy_session,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
| OP_MODIFIES_SOMETHING,
nfsd41: proc stubs Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:27:58 +08:00
.op_name = "OP_DESTROY_SESSION",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
nfsd41: proc stubs Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:27:58 +08:00
},
[OP_SEQUENCE] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_sequence,
nfsd41: enforce NFS4ERR_SEQUENCE_POS operation order rules for minorversion != 0 only. Signed-off-by: Andy Adamson<andros@netapp.com> [nfsd41: do not verify nfserr_sequence_pos for minorversion 0] Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:28:12 +08:00
.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP,
nfsd41: proc stubs Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:27:58 +08:00
.op_name = "OP_SEQUENCE",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_sequence_rsize,
nfsd41: proc stubs Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-03 13:27:58 +08:00
},
NFSD: allow OP_DESTROY_CLIENTID to be only op in COMPOUND DESTROY_CLIENTID MAY be preceded with a SEQUENCE operation as long as the client ID derived from the session ID of SEQUENCE is not the same as the client ID to be destroyed. If the client IDs are the same, then the server MUST return NFS4ERR_CLIENTID_BUSY. (that's not implemented yet) If DESTROY_CLIENTID is not prefixed by SEQUENCE, it MUST be the only operation in the COMPOUND request (otherwise, the server MUST return NFS4ERR_NOT_ONLY_OP). This fixes the error return; before, we returned NFS4ERR_OP_NOT_IN_SESSION; after this patch, we return NFS4ERR_NOTSUPP. Signed-off-by: Benny Halevy <benny@tonian.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-06-16 23:39:10 +08:00
[OP_DESTROY_CLIENTID] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_destroy_clientid,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
| OP_MODIFIES_SOMETHING,
NFSD: allow OP_DESTROY_CLIENTID to be only op in COMPOUND DESTROY_CLIENTID MAY be preceded with a SEQUENCE operation as long as the client ID derived from the session ID of SEQUENCE is not the same as the client ID to be destroyed. If the client IDs are the same, then the server MUST return NFS4ERR_CLIENTID_BUSY. (that's not implemented yet) If DESTROY_CLIENTID is not prefixed by SEQUENCE, it MUST be the only operation in the COMPOUND request (otherwise, the server MUST return NFS4ERR_NOT_ONLY_OP). This fixes the error return; before, we returned NFS4ERR_OP_NOT_IN_SESSION; after this patch, we return NFS4ERR_NOTSUPP. Signed-off-by: Benny Halevy <benny@tonian.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-06-16 23:39:10 +08:00
.op_name = "OP_DESTROY_CLIENTID",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
NFSD: allow OP_DESTROY_CLIENTID to be only op in COMPOUND DESTROY_CLIENTID MAY be preceded with a SEQUENCE operation as long as the client ID derived from the session ID of SEQUENCE is not the same as the client ID to be destroyed. If the client IDs are the same, then the server MUST return NFS4ERR_CLIENTID_BUSY. (that's not implemented yet) If DESTROY_CLIENTID is not prefixed by SEQUENCE, it MUST be the only operation in the COMPOUND request (otherwise, the server MUST return NFS4ERR_NOT_ONLY_OP). This fixes the error return; before, we returned NFS4ERR_OP_NOT_IN_SESSION; after this patch, we return NFS4ERR_NOTSUPP. Signed-off-by: Benny Halevy <benny@tonian.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-06-16 23:39:10 +08:00
},
nfsd4: implement reclaim_complete This is a mandatory operation. Also, here (not in open) is where we should be committing the reboot recovery information. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2010-04-20 03:11:28 +08:00
[OP_RECLAIM_COMPLETE] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_reclaim_complete,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_flags = ALLOWED_WITHOUT_FH | OP_MODIFIES_SOMETHING,
nfsd4: implement reclaim_complete This is a mandatory operation. Also, here (not in open) is where we should be committing the reboot recovery information. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2010-04-20 03:11:28 +08:00
.op_name = "OP_RECLAIM_COMPLETE",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
nfsd4: implement reclaim_complete This is a mandatory operation. Also, here (not in open) is where we should be committing the reboot recovery information. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2010-04-20 03:11:28 +08:00
},
nfsd4: implement secinfo_no_name Implementation of this operation is mandatory for NFSv4.1. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2010-12-16 22:51:13 +08:00
[OP_SECINFO_NO_NAME] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_secinfo_no_name,
nfsd4: define nfsd4_secinfo_no_name_release() Commit 34b1744c91cc ("nfsd4: define ->op_release for compound ops") defined a couple ->op_release functions and run them if necessary. But there's a problem with that is that it reused nfsd4_secinfo_release() as the op_release of OP_SECINFO_NO_NAME, and caused a leak on struct nfsd4_secinfo_no_name in nfsd4_encode_secinfo_no_name(), because there's no .si_exp field in struct nfsd4_secinfo_no_name. I found this because I was unable to umount an ext4 partition after exporting it via NFS & run fsstress on the nfs mount. A simplified reproducer would be: # mount a local-fs device at /mnt/test, and export it via NFS with # fsid=0 export option (this is required) mount /dev/sda5 /mnt/test echo "/mnt/test *(rw,no_root_squash,fsid=0)" >> /etc/exports service nfs restart # locally mount the nfs export with all default, note that I have # nfsv4.1 configured as the default nfs version, because of the # fsid export option, v4 mount would fail and fall back to v3 mount localhost:/mnt/test /mnt/nfs # try to umount the underlying device, but got EBUSY umount /mnt/nfs service nfs stop umount /mnt/test <=== EBUSY here Fixed it by defining a separate nfsd4_secinfo_no_name_release() function as the op_release method of OP_SECINFO_NO_NAME that releases the correct nfsd4_secinfo_no_name structure. Fixes: 34b1744c91cc ("nfsd4: define ->op_release for compound ops") Signed-off-by: Eryu Guan <eguan@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-09-29 15:01:10 +08:00
.op_release = nfsd4_secinfo_no_name_release,
nfsd4: fix wrongsec handling for PUTFH + op cases When PUTFH is followed by an operation that uses the filehandle, and when the current client is using a security flavor that is inconsistent with the given filehandle, we have a choice: we can return WRONGSEC either when the current filehandle is set using the PUTFH, or when the filehandle is first used by the following operation. Follow the recommendations of RFC 5661 in making this choice. (Our current behavior prevented the client from doing security negotiation by returning WRONGSEC on PUTFH+SECINFO_NO_NAME.) Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-04-09 05:00:50 +08:00
.op_flags = OP_HANDLES_WRONGSEC,
nfsd4: implement secinfo_no_name Implementation of this operation is mandatory for NFSv4.1. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2010-12-16 22:51:13 +08:00
.op_name = "OP_SECINFO_NO_NAME",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_secinfo_rsize,
nfsd4: implement secinfo_no_name Implementation of this operation is mandatory for NFSv4.1. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2010-12-16 22:51:13 +08:00
},
NFSD: Added TEST_STATEID operation This operation is used by the client to check the validity of a list of stateids. Signed-off-by: Bryan Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-13 22:50:48 +08:00
[OP_TEST_STATEID] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_test_stateid,
NFSD: Added TEST_STATEID operation This operation is used by the client to check the validity of a list of stateids. Signed-off-by: Bryan Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-13 22:50:48 +08:00
.op_flags = ALLOWED_WITHOUT_FH,
.op_name = "OP_TEST_STATEID",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_test_stateid_rsize,
NFSD: Added TEST_STATEID operation This operation is used by the client to check the validity of a list of stateids. Signed-off-by: Bryan Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-13 22:50:48 +08:00
},
NFSD: added FREE_STATEID operation This operation is used by the client to tell the server to free a stateid. Signed-off-by: Bryan Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-13 23:04:21 +08:00
[OP_FREE_STATEID] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_free_stateid,
nfsd41: try to check reply size before operation For checking the size of reply before calling a operation, we need try to get maxsize of the operation's reply. v3: using new method as Bruce said, "we could handle operations in two different ways: - For operations that actually change something (write, rename, open, close, ...), do it the way we're doing it now: be very careful to estimate the size of the response before even processing the operation. - For operations that don't change anything (read, getattr, ...) just go ahead and do the operation. If you realize after the fact that the response is too large, then return the error at that point. So we'd add another flag to op_flags: say, OP_MODIFIES_SOMETHING. And for operations with OP_MODIFIES_SOMETHING set, we'd do the first thing. For operations without it set, we'd do the second." Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com> [bfields@redhat.com: crash, don't attempt to handle, undefined op_rsize_bop] Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-28 18:18:56 +08:00
.op_flags = ALLOWED_WITHOUT_FH | OP_MODIFIES_SOMETHING,
NFSD: added FREE_STATEID operation This operation is used by the client to tell the server to free a stateid. Signed-off-by: Bryan Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-13 23:04:21 +08:00
.op_name = "OP_FREE_STATEID",
nfsd4: properly type op_get_currentstateid callbacks Pass union nfsd4_op_u to the op_set_currentstateid callbacks instead of using unsafe function pointer casts. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:37:33 +08:00
.op_get_currentstateid = nfsd4_get_freestateid,
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
NFSD: added FREE_STATEID operation This operation is used by the client to tell the server to free a stateid. Signed-off-by: Bryan Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-13 23:04:21 +08:00
},
nfsd: trivial GET_DIR_DELEGATION support This adds basic infrastructure for handing GET_DIR_DELEGATION calls from clients, including the decoders and encoders. For now, it always just returns NFS4_OK + GDD4_UNAVAIL. Eventually clients may start sending this operation, and it's better if we can return GDD4_UNAVAIL instead of having to abort the whole compound. Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-03-23 21:25:54 +08:00
[OP_GET_DIR_DELEGATION] = {
.op_func = nfsd4_get_dir_delegation,
.op_flags = OP_MODIFIES_SOMETHING,
.op_name = "OP_GET_DIR_DELEGATION",
.op_rsize_bop = nfsd4_get_dir_delegation_rsize,
},
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
#ifdef CONFIG_NFSD_PNFS
[OP_GETDEVICEINFO] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_getdeviceinfo,
nfsd4: define ->op_release for compound ops Run a separate ->op_release function if necessary instead of depending on the xdr encoder to do this. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-05-06 05:09:37 +08:00
.op_release = nfsd4_getdeviceinfo_release,
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
.op_flags = ALLOWED_WITHOUT_FH,
.op_name = "OP_GETDEVICEINFO",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_getdeviceinfo_rsize,
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
},
[OP_LAYOUTGET] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_layoutget,
nfsd4: define ->op_release for compound ops Run a separate ->op_release function if necessary instead of depending on the xdr encoder to do this. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-05-06 05:09:37 +08:00
.op_release = nfsd4_layoutget_release,
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
.op_flags = OP_MODIFIES_SOMETHING,
.op_name = "OP_LAYOUTGET",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_layoutget_rsize,
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
},
[OP_LAYOUTCOMMIT] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_layoutcommit,
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
.op_flags = OP_MODIFIES_SOMETHING,
.op_name = "OP_LAYOUTCOMMIT",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_layoutcommit_rsize,
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
},
[OP_LAYOUTRETURN] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_layoutreturn,
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
.op_flags = OP_MODIFIES_SOMETHING,
.op_name = "OP_LAYOUTRETURN",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_layoutreturn_rsize,
nfsd: implement pNFS operations Add support for the GETDEVICEINFO, LAYOUTGET, LAYOUTCOMMIT and LAYOUTRETURN NFSv4.1 operations, as well as backing code to manage outstanding layouts and devices. Layout management is very straight forward, with a nfs4_layout_stateid structure that extends nfs4_stid to manage layout stateids as the top-level structure. It is linked into the nfs4_file and nfs4_client structures like the other stateids, and contains a linked list of layouts that hang of the stateid. The actual layout operations are implemented in layout drivers that are not part of this commit, but will be added later. The worst part of this commit is the management of the pNFS device IDs, which suffers from a specification that is not sanely implementable due to the fact that the device-IDs are global and not bound to an export, and have a small enough size so that we can't store the fsid portion of a file handle, and must never be reused. As we still do need perform all export authentication and validation checks on a device ID passed to GETDEVICEINFO we are caught between a rock and a hard place. To work around this issue we add a new hash that maps from a 64-bit integer to a fsid so that we can look up the export to authenticate against it, a 32-bit integer as a generation that we can bump when changing the device, and a currently unused 32-bit integer that could be used in the future to handle more than a single device per export. Entries in this hash table are never deleted as we can't reuse the ids anyway, and would have a severe lifetime problem anyway as Linux export structures are temporary structures that can go away under load. Parts of the XDR data, structures and marshaling/unmarshaling code, as well as many concepts are derived from the old pNFS server implementation from Andy Adamson, Benny Halevy, Dean Hildebrand, Marc Eshel, Fred Isaman, Mike Sager, Ricardo Labiaga and many others. Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-05-05 19:11:59 +08:00
},
#endif /* CONFIG_NFSD_PNFS */
NFSD: Implement SEEK This patch adds server support for the NFS v4.2 operation SEEK, which returns the position of the next hole or data segment in a file. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-09-27 01:58:27 +08:00
/* NFSv4.2 operations */
nfsd: Add ALLOCATE support The ALLOCATE operation is used to preallocate space in a file. I can do this by using vfs_fallocate() to do the actual preallocation. ALLOCATE only returns a status indicator, so we don't need to write a special encode() function. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-11-08 03:44:26 +08:00
[OP_ALLOCATE] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_allocate,
NFSD remove OP_CACHEME from 4.2 op_flags OP_CACHEME is only for the 4.0 operations. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-12-05 03:09:21 +08:00
.op_flags = OP_MODIFIES_SOMETHING,
nfsd: Add ALLOCATE support The ALLOCATE operation is used to preallocate space in a file. I can do this by using vfs_fallocate() to do the actual preallocation. ALLOCATE only returns a status indicator, so we don't need to write a special encode() function. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-11-08 03:44:26 +08:00
.op_name = "OP_ALLOCATE",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
nfsd: Add ALLOCATE support The ALLOCATE operation is used to preallocate space in a file. I can do this by using vfs_fallocate() to do the actual preallocation. ALLOCATE only returns a status indicator, so we don't need to write a special encode() function. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-11-08 03:44:26 +08:00
},
nfsd: Add DEALLOCATE support DEALLOCATE only returns a status value, meaning we can use the noop() xdr encoder to reply to the client. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-11-08 03:44:27 +08:00
[OP_DEALLOCATE] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_deallocate,
NFSD remove OP_CACHEME from 4.2 op_flags OP_CACHEME is only for the 4.0 operations. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-12-05 03:09:21 +08:00
.op_flags = OP_MODIFIES_SOMETHING,
nfsd: Add DEALLOCATE support DEALLOCATE only returns a status value, meaning we can use the noop() xdr encoder to reply to the client. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-11-08 03:44:27 +08:00
.op_name = "OP_DEALLOCATE",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
nfsd: Add DEALLOCATE support DEALLOCATE only returns a status value, meaning we can use the noop() xdr encoder to reply to the client. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-11-08 03:44:27 +08:00
},
nfsd: implement the NFSv4.2 CLONE operation This is basically a remote version of the btrfs CLONE operation, so the implementation is fairly trivial. Made even more trivial by stealing the XDR code and general framework Anna Schumaker's COPY prototype. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 19:59:52 +08:00
[OP_CLONE] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_clone,
NFSD remove OP_CACHEME from 4.2 op_flags OP_CACHEME is only for the 4.0 operations. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-12-05 03:09:21 +08:00
.op_flags = OP_MODIFIES_SOMETHING,
nfsd: implement the NFSv4.2 CLONE operation This is basically a remote version of the btrfs CLONE operation, so the implementation is fairly trivial. Made even more trivial by stealing the XDR code and general framework Anna Schumaker's COPY prototype. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 19:59:52 +08:00
.op_name = "OP_CLONE",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_only_status_rsize,
nfsd: implement the NFSv4.2 CLONE operation This is basically a remote version of the btrfs CLONE operation, so the implementation is fairly trivial. Made even more trivial by stealing the XDR code and general framework Anna Schumaker's COPY prototype. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 19:59:52 +08:00
},
NFSD: Implement the COPY call I only implemented the sync version of this call, since it's the easiest. I can simply call vfs_copy_range() and have the vfs do the right thing for the filesystem being exported. Signed-off-by: Anna Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-09-08 03:57:30 +08:00
[OP_COPY] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_copy,
NFSD remove OP_CACHEME from 4.2 op_flags OP_CACHEME is only for the 4.0 operations. Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-12-05 03:09:21 +08:00
.op_flags = OP_MODIFIES_SOMETHING,
NFSD: Implement the COPY call I only implemented the sync version of this call, since it's the easiest. I can simply call vfs_copy_range() and have the vfs do the right thing for the filesystem being exported. Signed-off-by: Anna Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-09-08 03:57:30 +08:00
.op_name = "OP_COPY",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_copy_rsize,
NFSD: Implement the COPY call I only implemented the sync version of this call, since it's the easiest. I can simply call vfs_copy_range() and have the vfs do the right thing for the filesystem being exported. Signed-off-by: Anna Schumaker <bjschuma@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-09-08 03:57:30 +08:00
},
NFSD: Add READ_PLUS data support This patch adds READ_PLUS support for returning a single NFS4_CONTENT_DATA segment to the client. This is basically the same as the READ operation, only with the extra information about data segments. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2020-09-29 01:08:58 +08:00
[OP_READ_PLUS] = {
.op_func = nfsd4_read,
.op_release = nfsd4_read_release,
.op_name = "OP_READ_PLUS",
.op_rsize_bop = nfsd4_read_plus_rsize,
.op_get_currentstateid = nfsd4_get_readstateid,
},
NFSD: Implement SEEK This patch adds server support for the NFS v4.2 operation SEEK, which returns the position of the next hole or data segment in a file. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-09-27 01:58:27 +08:00
[OP_SEEK] = {
nfsd4: properly type op_func callbacks Pass union nfsd4_op_u to the op_func callbacks instead of using unsafe function pointer casts. It also adds two missing structures to struct nfsd4_op.u to facilitate this. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:58:35 +08:00
.op_func = nfsd4_seek,
NFSD: Implement SEEK This patch adds server support for the NFS v4.2 operation SEEK, which returns the position of the next hole or data segment in a file. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-09-27 01:58:27 +08:00
.op_name = "OP_SEEK",
nfsd4: remove nfsd4op_rsize Except for a lot of unnecessary casts this typedef only has one user, so remove the casts and expand it in struct nfsd4_operation. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 02:42:10 +08:00
.op_rsize_bop = nfsd4_seek_rsize,
NFSD: Implement SEEK This patch adds server support for the NFS v4.2 operation SEEK, which returns the position of the next hole or data segment in a file. Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-09-27 01:58:27 +08:00
},
NFSD OFFLOAD_STATUS xdr Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:18 +08:00
[OP_OFFLOAD_STATUS] = {
.op_func = nfsd4_offload_status,
.op_name = "OP_OFFLOAD_STATUS",
.op_rsize_bop = nfsd4_offload_status_rsize,
},
NFSD OFFLOAD_CANCEL xdr Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-07-21 06:19:19 +08:00
[OP_OFFLOAD_CANCEL] = {
.op_func = nfsd4_offload_cancel,
.op_flags = OP_MODIFIES_SOMETHING,
.op_name = "OP_OFFLOAD_CANCEL",
.op_rsize_bop = nfsd4_only_status_rsize,
},
NFSD COPY_NOTIFY xdr Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
2019-08-08 23:14:59 +08:00
[OP_COPY_NOTIFY] = {
.op_func = nfsd4_copy_notify,
.op_flags = OP_MODIFIES_SOMETHING,
.op_name = "OP_COPY_NOTIFY",
.op_rsize_bop = nfsd4_copy_notify_rsize,
},
nfsd: implement the xattr functions and en/decode logic Implement the main entry points for the *XATTR operations. Add functions to calculate the reply size for the user extended attribute operations, and implement the XDR encode / decode logic for these operations. Add the user extended attributes operations to nfsd4_ops. Signed-off-by: Frank van der Linden <fllinden@amazon.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-06-24 06:39:26 +08:00
[OP_GETXATTR] = {
.op_func = nfsd4_getxattr,
.op_name = "OP_GETXATTR",
.op_rsize_bop = nfsd4_getxattr_rsize,
},
[OP_SETXATTR] = {
.op_func = nfsd4_setxattr,
.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
.op_name = "OP_SETXATTR",
.op_rsize_bop = nfsd4_setxattr_rsize,
},
[OP_LISTXATTRS] = {
.op_func = nfsd4_listxattrs,
.op_name = "OP_LISTXATTRS",
.op_rsize_bop = nfsd4_listxattrs_rsize,
},
[OP_REMOVEXATTR] = {
.op_func = nfsd4_removexattr,
.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
.op_name = "OP_REMOVEXATTR",
.op_rsize_bop = nfsd4_removexattr_rsize,
},
[PATCH] knfsd: nfsd4: reorganize compound ops Define an op descriptor struct, use it to simplify nfsd4_proc_compound(). Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 16:35:38 +08:00
};
nfsd: implement machine credential support for some operations This addresses the conundrum referenced in RFC5661 18.35.3, and will allow clients to return state to the server using the machine credentials. The biggest part of the problem is that we need to allow the client to send a compound op with integrity/privacy on mounts that don't have it enabled. Add server support for properly decoding and using spo_must_enforce and spo_must_allow bits. Add support for machine credentials to be used for CLOSE, OPEN_DOWNGRADE, LOCKU, DELEGRETURN, and TEST/FREE STATEID. Implement a check so as to not throw WRONGSEC errors when these operations are used if integrity/privacy isn't turned on. Without this, Linux clients with credentials that expired while holding delegations were getting stuck in an endless loop. Signed-off-by: Andrew Elble <aweits@rit.edu> Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-06-16 00:52:09 +08:00
/**
* nfsd4_spo_must_allow - Determine if the compound op contains an
* operation that is allowed to be sent with machine credentials
*
* @rqstp: a pointer to the struct svc_rqst
*
* Checks to see if the compound contains a spo_must_allow op
* and confirms that it was sent with the proper machine creds.
*/
bool nfsd4_spo_must_allow(struct svc_rqst *rqstp)
{
struct nfsd4_compoundres *resp = rqstp->rq_resp;
struct nfsd4_compoundargs *argp = rqstp->rq_argp;
nfsd: remove redundant assignment to pointer 'this' The pointer 'this' is being initialized with a value that is never read and it is being updated later with a new value. The initialization is redundant and can be removed. Addresses-Coverity: ("Unused value") Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2021-05-13 23:16:39 +08:00
struct nfsd4_op *this;
nfsd: implement machine credential support for some operations This addresses the conundrum referenced in RFC5661 18.35.3, and will allow clients to return state to the server using the machine credentials. The biggest part of the problem is that we need to allow the client to send a compound op with integrity/privacy on mounts that don't have it enabled. Add server support for properly decoding and using spo_must_enforce and spo_must_allow bits. Add support for machine credentials to be used for CLOSE, OPEN_DOWNGRADE, LOCKU, DELEGRETURN, and TEST/FREE STATEID. Implement a check so as to not throw WRONGSEC errors when these operations are used if integrity/privacy isn't turned on. Without this, Linux clients with credentials that expired while holding delegations were getting stuck in an endless loop. Signed-off-by: Andrew Elble <aweits@rit.edu> Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-06-16 00:52:09 +08:00
struct nfsd4_compound_state *cstate = &resp->cstate;
struct nfs4_op_map *allow = &cstate->clp->cl_spo_must_allow;
u32 opiter;
if (!cstate->minorversion)
return false;
nfsd: fix comparison to bool warning Fixes coccicheck warning: fs/nfsd/nfs4proc.c:3234:5-29: WARNING: Comparison to bool Signed-off-by: Zheng Bin <zhengbin13@huawei.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2020-09-11 12:10:14 +08:00
if (cstate->spo_must_allowed)
nfsd: implement machine credential support for some operations This addresses the conundrum referenced in RFC5661 18.35.3, and will allow clients to return state to the server using the machine credentials. The biggest part of the problem is that we need to allow the client to send a compound op with integrity/privacy on mounts that don't have it enabled. Add server support for properly decoding and using spo_must_enforce and spo_must_allow bits. Add support for machine credentials to be used for CLOSE, OPEN_DOWNGRADE, LOCKU, DELEGRETURN, and TEST/FREE STATEID. Implement a check so as to not throw WRONGSEC errors when these operations are used if integrity/privacy isn't turned on. Without this, Linux clients with credentials that expired while holding delegations were getting stuck in an endless loop. Signed-off-by: Andrew Elble <aweits@rit.edu> Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2016-06-16 00:52:09 +08:00
return true;
opiter = resp->opcnt;
while (opiter < argp->opcnt) {
this = &argp->ops[opiter++];
if (test_bit(this->opnum, allow->u.longs) &&
cstate->clp->cl_mach_cred &&
nfsd4_mach_creds_match(cstate->clp, rqstp)) {
cstate->spo_must_allowed = true;
return true;
}
}
cstate->spo_must_allowed = false;
return false;
}
nfsd4: more precise nfsd4_max_reply It will turn out to be useful to have a more accurate estimate of reply size; so, piggyback on the existing op reply-size estimators. Also move nfsd4_max_reply to nfs4proc.c to get easier access to struct nfsd4_operation and friends. (Thanks to Christoph Hellwig for pointing out that simplification.) Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-12 03:39:13 +08:00
int nfsd4_max_reply(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
nfsd: fix oops on unsupported operation I'm hitting the BUG in nfsd4_max_reply() at fs/nfsd/nfs4proc.c:2495 when client sends an operation the server doesn't support. in nfsd4_max_reply() it checks for NULL rsize_bop but a non-supported operation wouldn't have that set. Cc: Kinglong Mee <kinglongmee@gmail.com> Fixes: 2282cd2c05e2 "NFSD: Get response size before operation..." Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-03-24 02:36:20 +08:00
if (op->opnum == OP_ILLEGAL || op->status == nfserr_notsupp)
nfsd4: more precise nfsd4_max_reply It will turn out to be useful to have a more accurate estimate of reply size; so, piggyback on the existing op reply-size estimators. Also move nfsd4_max_reply to nfs4proc.c to get easier access to struct nfsd4_operation and friends. (Thanks to Christoph Hellwig for pointing out that simplification.) Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-12 03:39:13 +08:00
return op_encode_hdr_size * sizeof(__be32);
NFSD: Get response size before operation for all RPCs NFSD usess PAGE_SIZE as the reply size estimate for RPCs which don't support op_rsize_bop(), A PAGE_SIZE (4096) is larger than many real response sizes, eg, access (op_encode_hdr_size + 2), seek (op_encode_hdr_size + 3). This patch just adds op_rsize_bop() for all RPCs getting response size. An overestimate is generally safe but the tighter estimates are probably better. Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-02-03 22:36:00 +08:00
BUG_ON(OPDESC(op)->op_rsize_bop == NULL);
return OPDESC(op)->op_rsize_bop(rqstp, op);
nfsd4: more precise nfsd4_max_reply It will turn out to be useful to have a more accurate estimate of reply size; so, piggyback on the existing op reply-size estimators. Also move nfsd4_max_reply to nfs4proc.c to get easier access to struct nfsd4_operation and friends. (Thanks to Christoph Hellwig for pointing out that simplification.) Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-12 03:39:13 +08:00
}
nfsd4: fix encoding of out-of-space replies If nfsd4_check_resp_size() returns an error then we should really be truncating the reply here, otherwise we may leave extra garbage at the end of the rpc reply. Also add a warning to catch any cases where our reply-size estimates may be wrong in the case of a non-idempotent operation. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-07 09:39:29 +08:00
void warn_on_nonidempotent_op(struct nfsd4_op *op)
{
if (OPDESC(op)->op_flags & OP_MODIFIES_SOMETHING) {
NFSD: Make nfsd4_ops::opnum a u32 Avoid passing a "pointer to int" argument to xdr_stream_decode_u32. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-11-23 01:49:52 +08:00
pr_err("unable to encode reply to nonidempotent op %u (%s)\n",
nfsd4: fix encoding of out-of-space replies If nfsd4_check_resp_size() returns an error then we should really be truncating the reply here, otherwise we may leave extra garbage at the end of the rpc reply. Also add a warning to catch any cases where our reply-size estimates may be wrong in the case of a non-idempotent operation. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2014-03-07 09:39:29 +08:00
op->opnum, nfsd4_op_name(op->opnum));
WARN_ON_ONCE(1);
}
}
[NFSD] uninline nfsd4_op_name() There doesn't seem to be a compelling reason why nfsd4_op_name() is marked as "inline": It's only used in a dprintk(), and as long as it has only one caller non-ancient gcc versions anyway inline it automatically. This patch fixes the following compile error with gcc 3.4: ... CC fs/nfsd/nfs4proc.o nfs4proc.c: In function `nfsd4_proc_compound': nfs4proc.c:854: sorry, unimplemented: inlining failed in call to nfs4proc.c:897: sorry, unimplemented: called from here make[3]: *** [fs/nfsd/nfs4proc.o] Error 1 Reported-by: Adrian Bunk <bunk@kernel.org> Signed-off-by: Adrian Bunk <bunk@kernel.org> [ Also made it "const char *" - Linus] Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-08-09 00:26:42 +08:00
static const char *nfsd4_op_name(unsigned opnum)
nfsd: dprint operation names Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-07-02 16:15:03 +08:00
{
if (opnum < ARRAY_SIZE(nfsd4_ops))
return nfsd4_ops[opnum].op_name;
return "unknown_operation";
}
sunrpc: mark all struct svc_procinfo instances as const struct svc_procinfo contains function pointers, and marking it as constant avoids it being able to be used as an attach vector for code injections. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-12 22:11:49 +08:00
static const struct svc_procedure nfsd_procedures4[2] = {
NFSv4: kill off complicated macro 'PROC' J. Bruce Fields wrote: ... > (This is extremely confusing code to track down: note that > proc->pc_decode is set to nfs4svc_decode_compoundargs() by the PROC() > macro at the end of fs/nfsd/nfs4proc.c. Which means, for example, that > grepping for nfs4svc_decode_compoundargs() gets you nowhere. Patches to > kill off that macro would be welcomed....) the macro 'PROC' is complicated and obscure, it had better be killed off in order to make the code more clear. Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-05-19 14:09:54 +08:00
[NFSPROC4_NULL] = {
nfsd: use named initializers in PROC() Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-08 23:59:13 +08:00
.pc_func = nfsd4_proc_null,
NFSD: Add common helpers to decode void args and encode void results Start off the conversion to xdr_stream by de-duplicating the functions that decode void arguments and encode void results. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-11-06 03:48:29 +08:00
.pc_decode = nfssvc_decode_voidarg,
.pc_encode = nfssvc_encode_voidres,
.pc_argsize = sizeof(struct nfsd_voidargs),
SUNRPC: Parametrize how much of argsize should be zeroed Currently, SUNRPC clears the whole of .pc_argsize before processing each incoming RPC transaction. Add an extra parameter to struct svc_procedure to enable upper layers to reduce the amount of each operation's argument structure that is zeroed by SUNRPC. The size of struct nfsd4_compoundargs, in particular, is a lot to clear on each incoming RPC Call. A subsequent patch will cut this down to something closer to what NFSv2 and NFSv3 uses. This patch should cause no behavior changes. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:22:38 +08:00
.pc_argzero = sizeof(struct nfsd_voidargs),
NFSD: Add common helpers to decode void args and encode void results Start off the conversion to xdr_stream by de-duplicating the functions that decode void arguments and encode void results. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-11-06 03:48:29 +08:00
.pc_ressize = sizeof(struct nfsd_voidres),
NFSv4: kill off complicated macro 'PROC' J. Bruce Fields wrote: ... > (This is extremely confusing code to track down: note that > proc->pc_decode is set to nfs4svc_decode_compoundargs() by the PROC() > macro at the end of fs/nfsd/nfs4proc.c. Which means, for example, that > grepping for nfs4svc_decode_compoundargs() gets you nowhere. Patches to > kill off that macro would be welcomed....) the macro 'PROC' is complicated and obscure, it had better be killed off in order to make the code more clear. Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-05-19 14:09:54 +08:00
.pc_cachetype = RC_NOCACHE,
.pc_xdrressize = 1,
SUNRPC: Make trace_svc_process() display the RPC procedure symbolically The next few patches will employ these strings to help make server- side trace logs more human-readable. A similar technique is already in use in kernel RPC client code. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-09-18 05:22:49 +08:00
.pc_name = "NULL",
NFSv4: kill off complicated macro 'PROC' J. Bruce Fields wrote: ... > (This is extremely confusing code to track down: note that > proc->pc_decode is set to nfs4svc_decode_compoundargs() by the PROC() > macro at the end of fs/nfsd/nfs4proc.c. Which means, for example, that > grepping for nfs4svc_decode_compoundargs() gets you nowhere. Patches to > kill off that macro would be welcomed....) the macro 'PROC' is complicated and obscure, it had better be killed off in order to make the code more clear. Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-05-19 14:09:54 +08:00
},
[NFSPROC4_COMPOUND] = {
nfsd: use named initializers in PROC() Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-08 23:59:13 +08:00
.pc_func = nfsd4_proc_compound,
sunrpc: properly type pc_decode callbacks Drop the argp argument as it can trivially be derived from the rqstp argument. With that all functions now have the same prototype, and we can remove the unsafe casting to kxdrproc_t. Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-05-09 01:01:48 +08:00
.pc_decode = nfs4svc_decode_compoundargs,
sunrpc: properly type pc_encode callbacks Drop the resp argument as it can trivially be derived from the rqstp argument. With that all functions now have the same prototype, and we can remove the unsafe casting to kxdrproc_t. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Trond Myklebust <trond.myklebust@primarydata.com>
2017-05-09 01:42:02 +08:00
.pc_encode = nfs4svc_encode_compoundres,
NFSv4: kill off complicated macro 'PROC' J. Bruce Fields wrote: ... > (This is extremely confusing code to track down: note that > proc->pc_decode is set to nfs4svc_decode_compoundargs() by the PROC() > macro at the end of fs/nfsd/nfs4proc.c. Which means, for example, that > grepping for nfs4svc_decode_compoundargs() gets you nowhere. Patches to > kill off that macro would be welcomed....) the macro 'PROC' is complicated and obscure, it had better be killed off in order to make the code more clear. Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-05-19 14:09:54 +08:00
.pc_argsize = sizeof(struct nfsd4_compoundargs),
NFSD: Reduce amount of struct nfsd4_compoundargs that needs clearing Have SunRPC clear everything except for the iops array. Then have each NFSv4 XDR decoder clear it's own argument before decoding. Now individual operations may have a large argument struct while not penalizing the vast majority of operations with a small struct. And, clearing the argument structure occurs as the argument fields are initialized, enabling the CPU to do write combining on that memory. In some cases, clearing is not even necessary because all of the fields in the argument structure are initialized by the decoder. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2022-09-13 05:22:44 +08:00
.pc_argzero = offsetof(struct nfsd4_compoundargs, iops),
NFSv4: kill off complicated macro 'PROC' J. Bruce Fields wrote: ... > (This is extremely confusing code to track down: note that > proc->pc_decode is set to nfs4svc_decode_compoundargs() by the PROC() > macro at the end of fs/nfsd/nfs4proc.c. Which means, for example, that > grepping for nfs4svc_decode_compoundargs() gets you nowhere. Patches to > kill off that macro would be welcomed....) the macro 'PROC' is complicated and obscure, it had better be killed off in order to make the code more clear. Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-05-19 14:09:54 +08:00
.pc_ressize = sizeof(struct nfsd4_compoundres),
nfsd4: call nfsd4_release_compoundargs from pc_release This simplifies cleanup a bit. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-17 05:15:10 +08:00
.pc_release = nfsd4_release_compoundargs,
NFSv4: kill off complicated macro 'PROC' J. Bruce Fields wrote: ... > (This is extremely confusing code to track down: note that > proc->pc_decode is set to nfs4svc_decode_compoundargs() by the PROC() > macro at the end of fs/nfsd/nfs4proc.c. Which means, for example, that > grepping for nfs4svc_decode_compoundargs() gets you nowhere. Patches to > kill off that macro would be welcomed....) the macro 'PROC' is complicated and obscure, it had better be killed off in order to make the code more clear. Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-05-19 14:09:54 +08:00
.pc_cachetype = RC_NOCACHE,
.pc_xdrressize = NFSD_BUFSIZE/4,
SUNRPC: Make trace_svc_process() display the RPC procedure symbolically The next few patches will employ these strings to help make server- side trace logs more human-readable. A similar technique is already in use in kernel RPC client code. Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2020-09-18 05:22:49 +08:00
.pc_name = "COMPOUND",
NFSv4: kill off complicated macro 'PROC' J. Bruce Fields wrote: ... > (This is extremely confusing code to track down: note that > proc->pc_decode is set to nfs4svc_decode_compoundargs() by the PROC() > macro at the end of fs/nfsd/nfs4proc.c. Which means, for example, that > grepping for nfs4svc_decode_compoundargs() gets you nowhere. Patches to > kill off that macro would be welcomed....) the macro 'PROC' is complicated and obscure, it had better be killed off in order to make the code more clear. Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-05-19 14:09:54 +08:00
},
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
};
SUNRPC: Use per-CPU counters to tally server RPC counts - Improves counting accuracy - Reduces cross-CPU memory traffic Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-01-10 23:31:54 +08:00
static DEFINE_PER_CPU_ALIGNED(unsigned long,
nfsd_count4[ARRAY_SIZE(nfsd_procedures4)]);
sunrpc: mark all struct svc_version instances as const Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Trond Myklebust <trond.myklebust@primarydata.com>
2017-05-12 22:21:37 +08:00
const struct svc_version nfsd_version4 = {
nfs/nfsd/sunrpc: enforce transport requirements for NFSv4 NFSv4 requires a transport "that is specified to avoid network congestion" (RFC 7530, section 3.1, paragraph 2). In practical terms, that means that you should not run NFSv4 over UDP. The server has never enforced that requirement, however. This patchset fixes this by adding a new flag to the svc_version that states that it has these transport requirements. With that, we can check that the transport has XPT_CONG_CTRL set before processing an RPC. If it doesn't we reject it with RPC_PROG_MISMATCH. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-02-25 02:25:24 +08:00
.vs_vers = 4,
SUNRPC: Use per-CPU counters to tally server RPC counts - Improves counting accuracy - Reduces cross-CPU memory traffic Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-01-10 23:31:54 +08:00
.vs_nproc = ARRAY_SIZE(nfsd_procedures4),
nfs/nfsd/sunrpc: enforce transport requirements for NFSv4 NFSv4 requires a transport "that is specified to avoid network congestion" (RFC 7530, section 3.1, paragraph 2). In practical terms, that means that you should not run NFSv4 over UDP. The server has never enforced that requirement, however. This patchset fixes this by adding a new flag to the svc_version that states that it has these transport requirements. With that, we can check that the transport has XPT_CONG_CTRL set before processing an RPC. If it doesn't we reject it with RPC_PROG_MISMATCH. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-02-25 02:25:24 +08:00
.vs_proc = nfsd_procedures4,
SUNRPC: Use per-CPU counters to tally server RPC counts - Improves counting accuracy - Reduces cross-CPU memory traffic Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2023-01-10 23:31:54 +08:00
.vs_count = nfsd_count4,
nfs/nfsd/sunrpc: enforce transport requirements for NFSv4 NFSv4 requires a transport "that is specified to avoid network congestion" (RFC 7530, section 3.1, paragraph 2). In practical terms, that means that you should not run NFSv4 over UDP. The server has never enforced that requirement, however. This patchset fixes this by adding a new flag to the svc_version that states that it has these transport requirements. With that, we can check that the transport has XPT_CONG_CTRL set before processing an RPC. If it doesn't we reject it with RPC_PROG_MISMATCH. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2017-02-25 02:25:24 +08:00
.vs_dispatch = nfsd_dispatch,
.vs_xdrsize = NFS4_SVC_XDRSIZE,
.vs_rpcb_optnl = true,
.vs_need_cong_ctrl = true,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
};
Reference in New Issue Copy Permalink