d/freebsd-nq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
44bc301921
freebsd-nq/sys/fs/smbfs/smbfs_io.c

678 lines
17 KiB
C
Raw Normal View History

/* -> /*- for copyright notices, minor format tweaks as necessary
2005-01-06 18:10:42 +00:00
/*-
sys/fs: further adoption of SPDX licensing ID tags. Mainly focus on files that use BSD 2-Clause license, however the tool I was using misidentified many licenses so this was mostly a manual - error prone - task. The Software Package Data Exchange (SPDX) group provides a specification to make it easier for automated tools to detect and summarize well known opensource licenses. We are gradually adopting the specification, noting that the tags are considered only advisory and do not, in any way, superceed or replace the license texts.
2017-11-27 15:15:37 +00:00
* SPDX-License-Identifier: BSD-2-Clause-FreeBSD
*
Switch to our preferred 2-clause BSD license. Approved by: bp
2010-04-07 16:50:38 +00:00
* Copyright (c) 2000-2001 Boris Popov
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
* All rights reserved.
*
* 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.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR 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.
*
* $FreeBSD$
*
*/
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/fcntl.h>
#include <sys/bio.h>
#include <sys/buf.h>
#include <sys/mount.h>
#include <sys/namei.h>
#include <sys/vnode.h>
#include <sys/dirent.h>
Change VM_OBJECT_LOCK/UNLOCK() -> VM_OBJECT_WLOCK/WUNLOCK() to reflect the recent switch of the vm object lock to a rwlock. Reported by: attilio
2013-05-04 14:27:28 +00:00
#include <sys/rwlock.h>
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
#include <sys/signalvar.h>
#include <sys/sysctl.h>
- Update the vmmeter statistics for vnode pageins and pageouts in getpages/putpages. - Use vm_page_undirty() instead of messing with pages' dirty fields directly.
2001-07-04 19:55:01 +00:00
#include <sys/vmmeter.h>
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
#include <vm/vm.h>
After the PHYS_TO_VM_PAGE() function was de-inlined, the main reason to pull vm_param.h was removed. Other big dependency of vm_page.h on vm_param.h are PA_LOCK* definitions, which are only needed for in-kernel code, because modules use KBI-safe functions to lock the pages. Stop including vm_param.h into vm_page.h. Include vm_param.h explicitely for the kernel code which needs it. Suggested and reviewed by: alc MFC after: 2 weeks
2012-08-05 14:11:42 +00:00
#include <vm/vm_param.h>
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
#include <vm/vm_page.h>
#include <vm/vm_extern.h>
#include <vm/vm_object.h>
#include <vm/vm_pager.h>
#include <vm/vnode_pager.h>
/*
#include <sys/ioccom.h>
*/
#include <netsmb/smb.h>
#include <netsmb/smb_conn.h>
#include <netsmb/smb_subr.h>
#include <fs/smbfs/smbfs.h>
#include <fs/smbfs/smbfs_node.h>
#include <fs/smbfs/smbfs_subr.h>
/*#define SMBFS_RWGENERIC*/
Allocate pager bufs from UMA instead of 80-ish mutex protected linked list. o In vm_pager_bufferinit() create pbuf_zone and start accounting on how many pbufs are we going to have set. In various subsystems that are going to utilize pbufs create private zones via call to pbuf_zsecond_create(). The latter calls uma_zsecond_create(), and sets a limit on created zone. After startup preallocate pbufs according to requirements of all pbuf zones. Subsystems that used to have a private limit with old allocator now have private pbuf zones: md(4), fusefs, NFS client, smbfs, VFS cluster, FFS, swap, vnode pager. The following subsystems use shared pbuf zone: cam(4), nvme(4), physio(9), aio(4). They should have their private limits, but changing that is out of scope of this commit. o Fetch tunable value of kern.nswbuf from init_param2() and while here move NSWBUF_MIN to opt_param.h and eliminate opt_swap.h, that was holding only this option. Default values aren't touched by this commit, but they probably should be reviewed wrt to modern hardware. This change removes a tight bottleneck from sendfile(2) operation, that uses pbufs in vnode pager. Other pagers also would benefit from faster allocation. Together with: gallatin Tested by: pho
2019-01-15 01:02:16 +00:00
extern uma_zone_t smbfs_pbuf_zone;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
static int smbfs_fastlookup = 1;
SYSCTL_DECL(_vfs_smbfs);
SYSCTL_INT(_vfs_smbfs, OID_AUTO, fastlookup, CTLFLAG_RW, &smbfs_fastlookup, 0, "");
#define DE_SIZE (sizeof(struct dirent))
static int
smbfs_readvdir(struct vnode *vp, struct uio *uio, struct ucred *cred)
{
struct dirent de;
struct componentname cn;
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
struct smb_cred *scred;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
struct smbfs_fctx *ctx;
struct vnode *newvp;
struct smbnode *np = VTOSMB(vp);
int error/*, *eofflag = ap->a_eofflag*/;
long offset, limit;
np = VTOSMB(vp);
SMBVDEBUG("dirname='%s'\n", np->n_name);
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
scred = smbfs_malloc_scred();
smb_makescred(scred, uio->uio_td, cred);
Clean up whitespace, s/register //, refrain from strong urge to ANSIfy.
2003-03-02 15:56:49 +00:00
offset = uio->uio_offset / DE_SIZE; /* offset in the directory */
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
limit = uio->uio_resid / DE_SIZE;
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
if (uio->uio_resid < DE_SIZE || uio->uio_offset < 0) {
error = EINVAL;
goto out;
}
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
while (limit && offset < 2) {
limit--;
bzero((caddr_t)&de, DE_SIZE);
de.d_reclen = DE_SIZE;
de.d_fileno = (offset == 0) ? np->n_ino :
Until now, smbfs_fullpath() computed the full path starting from the vnode and following back the chain of n_parent pointers up to the root, without acquiring the locks of the n_parent vnodes analyzed during the computation. This is immediately wrong because if the vnode lock is not held there's no guarantee on the validity of the vnode pointer or the data. In order to fix, store the whole path in the smbnode structure so that smbfs_fullpath() can use this information. Discussed with: kib Reported and tested by: pho Sponsored by: iXsystems inc.
2012-11-22 08:58:29 +00:00
(np->n_parent ? np->n_parentino : 2);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (de.d_fileno == 0)
de.d_fileno = 0x7ffffffd + offset;
Ensure that dirent's d_off field is initialized We have the d_off field in struct dirent for providing the seek offset of the next directory entry. Several filesystems were not initializing the field, which ends up being copied out to userland. Reported by: Syed Faraz Abrar <faraz@elttam.com> Reviewed by: kib MFC after: 3 days Sponsored by: The FreeBSD Foundation Differential Revision: https://reviews.freebsd.org/D27792
2021-01-03 16:32:30 +00:00
de.d_off = offset + 1;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
de.d_namlen = offset + 1;
de.d_name[0] = '.';
de.d_name[1] = '.';
de.d_type = DT_DIR;
Ensure that directory entry padding bytes are zeroed. Directory entries must be padded to maintain alignment; in many filesystems the padding was not initialized, resulting in stack memory being copied out to userspace. With the ino64 work there are also some explicit pad fields in struct dirent. Add a subroutine to clear these bytes and use it in the in-tree filesystems. The NFS client is omitted for now as it was fixed separately in r340787. Reported by: Thomas Barabosch, Fraunhofer FKIE Reviewed by: kib MFC after: 3 days Sponsored by: The FreeBSD Foundation
2018-11-23 22:24:59 +00:00
dirent_terminate(&de);
uiomove-related caddr_t -> void * (just the low-hanging fruit)
2003-03-02 15:50:23 +00:00
error = uiomove(&de, DE_SIZE, uio);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (error)
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
goto out;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
offset++;
uio->uio_offset += DE_SIZE;
}
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
if (limit == 0) {
error = 0;
goto out;
}
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (offset != np->n_dirofs || np->n_dirseq == NULL) {
SMBVDEBUG("Reopening search %ld:%ld\n", offset, np->n_dirofs);
if (np->n_dirseq) {
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
smbfs_findclose(np->n_dirseq, scred);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
np->n_dirseq = NULL;
}
np->n_dirofs = 2;
error = smbfs_findopen(np, "*", 1,
SMB_FA_SYSTEM | SMB_FA_HIDDEN | SMB_FA_DIR,
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
scred, &ctx);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (error) {
SMBVDEBUG("can not open search, error = %d", error);
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
goto out;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
}
np->n_dirseq = ctx;
} else
ctx = np->n_dirseq;
while (np->n_dirofs < offset) {
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
error = smbfs_findnext(ctx, offset - np->n_dirofs++, scred);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (error) {
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
smbfs_findclose(np->n_dirseq, scred);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
np->n_dirseq = NULL;
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
error = ENOENT ? 0 : error;
goto out;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
}
}
error = 0;
for (; limit; limit--, offset++) {
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
error = smbfs_findnext(ctx, limit, scred);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (error)
break;
np->n_dirofs++;
bzero((caddr_t)&de, DE_SIZE);
de.d_reclen = DE_SIZE;
de.d_fileno = ctx->f_attr.fa_ino;
Ensure that dirent's d_off field is initialized We have the d_off field in struct dirent for providing the seek offset of the next directory entry. Several filesystems were not initializing the field, which ends up being copied out to userland. Reported by: Syed Faraz Abrar <faraz@elttam.com> Reviewed by: kib MFC after: 3 days Sponsored by: The FreeBSD Foundation Differential Revision: https://reviews.freebsd.org/D27792
2021-01-03 16:32:30 +00:00
de.d_off = offset + 1;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
de.d_type = (ctx->f_attr.fa_attr & SMB_FA_DIR) ? DT_DIR : DT_REG;
de.d_namlen = ctx->f_nmlen;
bcopy(ctx->f_name, de.d_name, de.d_namlen);
Ensure that directory entry padding bytes are zeroed. Directory entries must be padded to maintain alignment; in many filesystems the padding was not initialized, resulting in stack memory being copied out to userspace. With the ino64 work there are also some explicit pad fields in struct dirent. Add a subroutine to clear these bytes and use it in the in-tree filesystems. The NFS client is omitted for now as it was fixed separately in r340787. Reported by: Thomas Barabosch, Fraunhofer FKIE Reviewed by: kib MFC after: 3 days Sponsored by: The FreeBSD Foundation
2018-11-23 22:24:59 +00:00
dirent_terminate(&de);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (smbfs_fastlookup) {
error = smbfs_nget(vp->v_mount, vp, ctx->f_name,
ctx->f_nmlen, &ctx->f_attr, &newvp);
if (!error) {
cn.cn_nameptr = de.d_name;
cn.cn_namelen = de.d_namlen;
Clean up whitespace, s/register //, refrain from strong urge to ANSIfy.
2003-03-02 15:56:49 +00:00
cache_enter(vp, newvp, &cn);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
vput(newvp);
}
}
uiomove-related caddr_t -> void * (just the low-hanging fruit)
2003-03-02 15:50:23 +00:00
error = uiomove(&de, DE_SIZE, uio);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (error)
break;
}
if (error == ENOENT)
error = 0;
uio->uio_offset = offset * DE_SIZE;
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
out:
smbfs_free_scred(scred);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
return error;
}
int
smbfs_readvnode(struct vnode *vp, struct uio *uiop, struct ucred *cred)
{
struct smbmount *smp = VFSTOSMBFS(vp->v_mount);
struct smbnode *np = VTOSMB(vp);
Catch up with KSE changes. Submitted by: Max Khon <fjoe@iclub.nsu.ru>
2001-12-02 08:56:58 +00:00
struct thread *td;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
struct vattr vattr;
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
struct smb_cred *scred;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
int error, lks;
UIO_NOCOPY is not supported for now, so refuse read opeartion if this flag is set. The full emulation of bio are on its way...
2002-04-26 03:49:02 +00:00
/*
* Protect against method which is not supported for now
*/
if (uiop->uio_segflg == UIO_NOCOPY)
return EOPNOTSUPP;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (vp->v_type != VREG && vp->v_type != VDIR) {
SMBFSERR("vn types other than VREG or VDIR are unsupported !\n");
return EIO;
}
if (uiop->uio_resid == 0)
return 0;
if (uiop->uio_offset < 0)
return EINVAL;
/* if (uiop->uio_offset + uiop->uio_resid > smp->nm_maxfilesize)
return EFBIG;*/
Catch up with KSE changes. Submitted by: Max Khon <fjoe@iclub.nsu.ru>
2001-12-02 08:56:58 +00:00
td = uiop->uio_td;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (vp->v_type == VDIR) {
Axe the 'thread' argument from VOP_ISLOCKED() and lockstatus() as it is always curthread. As KPI gets broken by this patch, manpages and __FreeBSD_version will be updated by further commits. Tested by: Andrea Barberio <insomniac at slackware dot it>
2008-02-25 18:45:57 +00:00
lks = LK_EXCLUSIVE; /* lockstatus(vp->v_vnlock); */
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (lks == LK_SHARED)
vn_lock() is currently only used with the 'curthread' passed as argument. Remove this argument and pass curthread directly to underlying VOP_LOCK1() VFS method. This modify makes the code cleaner and in particular remove an annoying dependence helping next lockmgr() cleanup. KPI results, obviously, changed. Manpage and FreeBSD_version will be updated through further commits. As a side note, would be valuable to say that next commits will address a similar cleanup about VFS methods, in particular vop_lock1 and vop_unlock. Tested by: Diego Sardina <siarodx at gmail dot com>, Andrea Di Pasquale <whyx dot it at gmail dot com>
2008-01-10 01:10:58 +00:00
vn_lock(vp, LK_UPGRADE | LK_RETRY);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
error = smbfs_readvdir(vp, uiop, cred);
if (lks == LK_SHARED)
vn_lock() is currently only used with the 'curthread' passed as argument. Remove this argument and pass curthread directly to underlying VOP_LOCK1() VFS method. This modify makes the code cleaner and in particular remove an annoying dependence helping next lockmgr() cleanup. KPI results, obviously, changed. Manpage and FreeBSD_version will be updated through further commits. As a side note, would be valuable to say that next commits will address a similar cleanup about VFS methods, in particular vop_lock1 and vop_unlock. Tested by: Diego Sardina <siarodx at gmail dot com>, Andrea Di Pasquale <whyx dot it at gmail dot com>
2008-01-10 01:10:58 +00:00
vn_lock(vp, LK_DOWNGRADE | LK_RETRY);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
return error;
}
/* biosize = SSTOCN(smp->sm_share)->sc_txmax;*/
if (np->n_flag & NMODIFIED) {
smbfs_attr_cacheremove(vp);
Decontextualize the couplet VOP_GETATTR / VOP_SETATTR as the passed thread was always curthread and totally unuseful. Tested by: Giovanni Trematerra <giovanni dot trematerra at gmail dot com>
2008-08-28 15:23:18 +00:00
error = VOP_GETATTR(vp, &vattr, cred);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (error)
return error;
np->n_mtime.tv_sec = vattr.va_mtime.tv_sec;
} else {
Decontextualize the couplet VOP_GETATTR / VOP_SETATTR as the passed thread was always curthread and totally unuseful. Tested by: Giovanni Trematerra <giovanni dot trematerra at gmail dot com>
2008-08-28 15:23:18 +00:00
error = VOP_GETATTR(vp, &vattr, cred);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (error)
return error;
if (np->n_mtime.tv_sec != vattr.va_mtime.tv_sec) {
Eliminate unused and constant arguments to smbfs_vinvalbuf()
2005-01-14 08:52:55 +00:00
error = smbfs_vinvalbuf(vp, td);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (error)
return error;
np->n_mtime.tv_sec = vattr.va_mtime.tv_sec;
}
}
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
scred = smbfs_malloc_scred();
smb_makescred(scred, td, cred);
error = smb_read(smp->sm_share, np->n_fid, uiop, scred);
smbfs_free_scred(scred);
return (error);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
}
int
smbfs_writevnode(struct vnode *vp, struct uio *uiop,
struct ucred *cred, int ioflag)
{
struct smbmount *smp = VTOSMBFS(vp);
struct smbnode *np = VTOSMB(vp);
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
struct smb_cred *scred;
Catch up with KSE changes. Submitted by: Max Khon <fjoe@iclub.nsu.ru>
2001-12-02 08:56:58 +00:00
struct thread *td;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
int error = 0;
if (vp->v_type != VREG) {
SMBERROR("vn types other than VREG unsupported !\n");
return EIO;
}
Change the code to use %jd as printf() placeholder for uio_offset and cast to intmax_t. Suggested by: pjd Sponsored by: iXsystems inc.
2012-10-31 02:54:44 +00:00
SMBVDEBUG("ofs=%jd,resid=%zd\n", (intmax_t)uiop->uio_offset,
uiop->uio_resid);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (uiop->uio_offset < 0)
return EINVAL;
/* if (uiop->uio_offset + uiop->uio_resid > smp->nm_maxfilesize)
return (EFBIG);*/
Catch up with KSE changes. Submitted by: Max Khon <fjoe@iclub.nsu.ru>
2001-12-02 08:56:58 +00:00
td = uiop->uio_td;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (ioflag & (IO_APPEND | IO_SYNC)) {
if (np->n_flag & NMODIFIED) {
smbfs_attr_cacheremove(vp);
Eliminate unused and constant arguments to smbfs_vinvalbuf()
2005-01-14 08:52:55 +00:00
error = smbfs_vinvalbuf(vp, td);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (error)
return error;
}
if (ioflag & IO_APPEND) {
Fix -Wundef from compiling the amd64 LINT.
2005-12-04 10:06:06 +00:00
#ifdef notyet
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
/*
* File size can be changed by another client
*/
smbfs_attr_cacheremove(vp);
Decontextualize the couplet VOP_GETATTR / VOP_SETATTR as the passed thread was always curthread and totally unuseful. Tested by: Giovanni Trematerra <giovanni dot trematerra at gmail dot com>
2008-08-28 15:23:18 +00:00
error = VOP_GETATTR(vp, &vattr, cred);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (error) return (error);
#endif
uiop->uio_offset = np->n_size;
}
}
if (uiop->uio_resid == 0)
return 0;
Move checking against RLIMIT_FSIZE into one place, vn_rlimit_fsize(). Reviewed by: kib
2010-05-05 16:44:25 +00:00
if (vn_rlimit_fsize(vp, uiop, td))
return (EFBIG);
fs: clean up empty lines in .c and .h files
2020-09-01 21:18:40 +00:00
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
scred = smbfs_malloc_scred();
smb_makescred(scred, td, cred);
error = smb_write(smp->sm_share, np->n_fid, uiop, scred);
smbfs_free_scred(scred);
Change the code to use %jd as printf() placeholder for uio_offset and cast to intmax_t. Suggested by: pjd Sponsored by: iXsystems inc.
2012-10-31 02:54:44 +00:00
SMBVDEBUG("after: ofs=%jd,resid=%zd\n", (intmax_t)uiop->uio_offset,
uiop->uio_resid);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (!error) {
if (uiop->uio_offset > np->n_size) {
np->n_size = uiop->uio_offset;
vnode_pager_setsize(vp, np->n_size);
}
}
return error;
}
/*
* Do an I/O operation to/from a cache block.
*/
int
Explicitly pass vnode to smbfs_doio() function.
2004-09-07 08:53:28 +00:00
smbfs_doio(struct vnode *vp, struct buf *bp, struct ucred *cr, struct thread *td)
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
{
struct smbmount *smp = VFSTOSMBFS(vp->v_mount);
struct smbnode *np = VTOSMB(vp);
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
struct uio *uiop;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
struct iovec io;
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
struct smb_cred *scred;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
int error = 0;
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
uiop = malloc(sizeof(struct uio), M_SMBFSDATA, M_WAITOK);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
uiop->uio_iov = &io;
uiop->uio_iovcnt = 1;
uiop->uio_segflg = UIO_SYSSPACE;
Catch up with KSE changes. Submitted by: Max Khon <fjoe@iclub.nsu.ru>
2001-12-02 08:56:58 +00:00
uiop->uio_td = td;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
scred = smbfs_malloc_scred();
smb_makescred(scred, td, cr);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (bp->b_iocmd == BIO_READ) {
io.iov_len = uiop->uio_resid = bp->b_bcount;
io.iov_base = bp->b_data;
uiop->uio_rw = UIO_READ;
switch (vp->v_type) {
case VREG:
uiop->uio_offset = ((off_t)bp->b_blkno) * DEV_BSIZE;
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
error = smb_read(smp->sm_share, np->n_fid, uiop, scred);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (error)
break;
if (uiop->uio_resid) {
int left = uiop->uio_resid;
int nread = bp->b_bcount - left;
if (left > 0)
bzero((char *)bp->b_data + nread, left);
}
break;
default:
printf("smbfs_doio: type %x unexpected\n",vp->v_type);
break;
Cleanup unnecessary semicolons from the kernel. Found with devel/coccinelle.
2016-04-10 23:07:00 +00:00
}
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (error) {
bp->b_error = error;
bp->b_ioflags |= BIO_ERROR;
}
} else { /* write */
if (((bp->b_blkno * DEV_BSIZE) + bp->b_dirtyend) > np->n_size)
bp->b_dirtyend = np->n_size - (bp->b_blkno * DEV_BSIZE);
if (bp->b_dirtyend > bp->b_dirtyoff) {
io.iov_len = uiop->uio_resid = bp->b_dirtyend - bp->b_dirtyoff;
uiop->uio_offset = ((off_t)bp->b_blkno) * DEV_BSIZE + bp->b_dirtyoff;
io.iov_base = (char *)bp->b_data + bp->b_dirtyoff;
uiop->uio_rw = UIO_WRITE;
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
error = smb_write(smp->sm_share, np->n_fid, uiop, scred);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
/*
* For an interrupted write, the buffer is still valid
* and the write hasn't been pushed to the server yet,
* so we can't set BIO_ERROR and report the interruption
* by setting B_EINTR. For the B_ASYNC case, B_EINTR
* is not relevant, so the rpc attempt is essentially
* a noop. For the case of a V3 write rpc not being
* committed to stable storage, the block is still
* dirty and requires either a commit rpc or another
* write rpc with iomode == NFSV3WRITE_FILESYNC before
* the block is reused. This is indicated by setting
* the B_DELWRI and B_NEEDCOMMIT flags.
*/
Clean up whitespace, s/register //, refrain from strong urge to ANSIfy.
2003-03-02 15:56:49 +00:00
if (error == EINTR
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
|| (!error && (bp->b_flags & B_NEEDCOMMIT))) {
bp->b_flags &= ~(B_INVAL|B_NOCACHE);
if ((bp->b_flags & B_ASYNC) == 0)
bp->b_flags |= B_EINTR;
if ((bp->b_flags & B_PAGING) == 0) {
bdirty(bp);
bp->b_flags &= ~B_DONE;
}
if ((bp->b_flags & B_ASYNC) == 0)
bp->b_flags |= B_EINTR;
Clean up whitespace, s/register //, refrain from strong urge to ANSIfy.
2003-03-02 15:56:49 +00:00
} else {
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (error) {
bp->b_ioflags |= BIO_ERROR;
bp->b_error = error;
}
bp->b_dirtyoff = bp->b_dirtyend = 0;
}
} else {
bp->b_resid = 0;
bufdone(bp);
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
free(uiop, M_SMBFSDATA);
smbfs_free_scred(scred);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
return 0;
}
}
bp->b_resid = uiop->uio_resid;
bufdone(bp);
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
free(uiop, M_SMBFSDATA);
smbfs_free_scred(scred);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
return error;
}
/*
* Vnode op for VM getpages.
* Wish wish .... get rid from multiple IO routines
*/
int
smbfs_getpages(ap)
struct vop_getpages_args /* {
struct vnode *a_vp;
vm_page_t *a_m;
int a_count;
int a_reqpage;
} */ *ap;
{
#ifdef SMBFS_RWGENERIC
Implement vop_std{get|put}pages() and add them to the default vop[]. Un-copy&paste all the VOP_{GET|PUT}PAGES() functions which do nothing but the default.
2001-05-01 08:34:45 +00:00
return vop_stdgetpages(ap);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
#else
A change to KPI of vm_pager_get_pages() and underlying VOP_GETPAGES(). o With new KPI consumers can request contiguous ranges of pages, and unlike before, all pages will be kept busied on return, like it was done before with the 'reqpage' only. Now the reqpage goes away. With new interface it is easier to implement code protected from race conditions. Such arrayed requests for now should be preceeded by a call to vm_pager_haspage() to make sure that request is possible. This could be improved later, making vm_pager_haspage() obsolete. Strenghtening the promises on the business of the array of pages allows us to remove such hacks as swp_pager_free_nrpage() and vm_pager_free_nonreq(). o New KPI accepts two integer pointers that may optionally point at values for read ahead and read behind, that a pager may do, if it can. These pages are completely owned by pager, and not controlled by the caller. This shifts the UFS-specific readahead logic from vm_fault.c, which should be file system agnostic, into vnode_pager.c. It also removes one VOP_BMAP() request per hard fault. Discussed with: kib, alc, jeff, scottl Sponsored by: Nginx, Inc. Sponsored by: Netflix
2015-12-16 21:30:45 +00:00
int i, error, nextoff, size, toff, npages, count;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
struct uio uio;
struct iovec iov;
vm_offset_t kva;
struct buf *bp;
struct vnode *vp;
Catch up with KSE changes. Submitted by: Max Khon <fjoe@iclub.nsu.ru>
2001-12-02 08:56:58 +00:00
struct thread *td;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
struct ucred *cred;
struct smbmount *smp;
struct smbnode *np;
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
struct smb_cred *scred;
Lock the vm object when freeing a vm page.
2003-06-19 03:38:05 +00:00
vm_object_t object;
A change to KPI of vm_pager_get_pages() and underlying VOP_GETPAGES(). o With new KPI consumers can request contiguous ranges of pages, and unlike before, all pages will be kept busied on return, like it was done before with the 'reqpage' only. Now the reqpage goes away. With new interface it is easier to implement code protected from race conditions. Such arrayed requests for now should be preceeded by a call to vm_pager_haspage() to make sure that request is possible. This could be improved later, making vm_pager_haspage() obsolete. Strenghtening the promises on the business of the array of pages allows us to remove such hacks as swp_pager_free_nrpage() and vm_pager_free_nonreq(). o New KPI accepts two integer pointers that may optionally point at values for read ahead and read behind, that a pager may do, if it can. These pages are completely owned by pager, and not controlled by the caller. This shifts the UFS-specific readahead logic from vm_fault.c, which should be file system agnostic, into vnode_pager.c. It also removes one VOP_BMAP() request per hard fault. Discussed with: kib, alc, jeff, scottl Sponsored by: Nginx, Inc. Sponsored by: Netflix
2015-12-16 21:30:45 +00:00
vm_page_t *pages;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
vp = ap->a_vp;
Lock the vm object when freeing a vm page.
2003-06-19 03:38:05 +00:00
if ((object = vp->v_object) == NULL) {
Track nfs's getpages() changes: Properly count v_vnodepgsin. Do not reread page if is already valid. Properly handle partially filled pages.
2002-04-23 14:30:43 +00:00
printf("smbfs_getpages: called with non-merged cache vnode??\n");
return VM_PAGER_ERROR;
}
Catch up with KSE changes. Submitted by: Max Khon <fjoe@iclub.nsu.ru>
2001-12-02 08:56:58 +00:00
td = curthread; /* XXX */
Simple p_ucred -> td_ucred changes to start using the per-thread ucred reference.
2002-02-27 18:32:23 +00:00
cred = td->td_ucred; /* XXX */
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
np = VTOSMB(vp);
smp = VFSTOSMBFS(vp->v_mount);
pages = ap->a_m;
Fix breakage caused by r292373 in ZFS/FUSE/NFS/SMBFS. With the new VOP_GETPAGES() KPI the "count" argument counts pages already, and doesn't need to be translated from bytes to pages. While here make it consistent that *rbehind and *rahead are updated only if we doesn't return error. Pointy hat to: glebius
2015-12-16 23:48:50 +00:00
npages = ap->a_count;
Track nfs's getpages() changes: Properly count v_vnodepgsin. Do not reread page if is already valid. Properly handle partially filled pages.
2002-04-23 14:30:43 +00:00
/*
* If the requested page is partially valid, just return it and
* allow the pager to zero-out the blanks. Partially valid pages
* can only occur at the file EOF.
A change to KPI of vm_pager_get_pages() and underlying VOP_GETPAGES(). o With new KPI consumers can request contiguous ranges of pages, and unlike before, all pages will be kept busied on return, like it was done before with the 'reqpage' only. Now the reqpage goes away. With new interface it is easier to implement code protected from race conditions. Such arrayed requests for now should be preceeded by a call to vm_pager_haspage() to make sure that request is possible. This could be improved later, making vm_pager_haspage() obsolete. Strenghtening the promises on the business of the array of pages allows us to remove such hacks as swp_pager_free_nrpage() and vm_pager_free_nonreq(). o New KPI accepts two integer pointers that may optionally point at values for read ahead and read behind, that a pager may do, if it can. These pages are completely owned by pager, and not controlled by the caller. This shifts the UFS-specific readahead logic from vm_fault.c, which should be file system agnostic, into vnode_pager.c. It also removes one VOP_BMAP() request per hard fault. Discussed with: kib, alc, jeff, scottl Sponsored by: Nginx, Inc. Sponsored by: Netflix
2015-12-16 21:30:45 +00:00
*
* XXXGL: is that true for SMB filesystem?
Track nfs's getpages() changes: Properly count v_vnodepgsin. Do not reread page if is already valid. Properly handle partially filled pages.
2002-04-23 14:30:43 +00:00
*/
Change VM_OBJECT_LOCK/UNLOCK() -> VM_OBJECT_WLOCK/WUNLOCK() to reflect the recent switch of the vm object lock to a rwlock. Reported by: attilio
2013-05-04 14:27:28 +00:00
VM_OBJECT_WLOCK(object);
(4/6) Protect page valid with the busy lock. Atomics are used for page busy and valid state when the shared busy is held. The details of the locking protocol and valid and dirty synchronization are in the updated vm_page.h comments. Reviewed by: kib, markj Tested by: pho Sponsored by: Netflix, Intel Differential Revision: https://reviews.freebsd.org/D21594
2019-10-15 03:45:41 +00:00
if (!vm_page_none_valid(pages[npages - 1]) && --npages == 0)
Fix breakage caused by r292373 in ZFS/FUSE/NFS/SMBFS. With the new VOP_GETPAGES() KPI the "count" argument counts pages already, and doesn't need to be translated from bytes to pages. While here make it consistent that *rbehind and *rahead are updated only if we doesn't return error. Pointy hat to: glebius
2015-12-16 23:48:50 +00:00
goto out;
Change VM_OBJECT_LOCK/UNLOCK() -> VM_OBJECT_WLOCK/WUNLOCK() to reflect the recent switch of the vm object lock to a rwlock. Reported by: attilio
2013-05-04 14:27:28 +00:00
VM_OBJECT_WUNLOCK(object);
Track nfs's getpages() changes: Properly count v_vnodepgsin. Do not reread page if is already valid. Properly handle partially filled pages.
2002-04-23 14:30:43 +00:00
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
scred = smbfs_malloc_scred();
smb_makescred(scred, td, cred);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
Allocate pager bufs from UMA instead of 80-ish mutex protected linked list. o In vm_pager_bufferinit() create pbuf_zone and start accounting on how many pbufs are we going to have set. In various subsystems that are going to utilize pbufs create private zones via call to pbuf_zsecond_create(). The latter calls uma_zsecond_create(), and sets a limit on created zone. After startup preallocate pbufs according to requirements of all pbuf zones. Subsystems that used to have a private limit with old allocator now have private pbuf zones: md(4), fusefs, NFS client, smbfs, VFS cluster, FFS, swap, vnode pager. The following subsystems use shared pbuf zone: cam(4), nvme(4), physio(9), aio(4). They should have their private limits, but changing that is out of scope of this commit. o Fetch tunable value of kern.nswbuf from init_param2() and while here move NSWBUF_MIN to opt_param.h and eliminate opt_swap.h, that was holding only this option. Default values aren't touched by this commit, but they probably should be reviewed wrt to modern hardware. This change removes a tight bottleneck from sendfile(2) operation, that uses pbufs in vnode pager. Other pagers also would benefit from faster allocation. Together with: gallatin Tested by: pho
2019-01-15 01:02:16 +00:00
bp = uma_zalloc(smbfs_pbuf_zone, M_WAITOK);
Get rid from extra #ifdefs.
2002-04-23 13:55:14 +00:00
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
kva = (vm_offset_t) bp->b_data;
pmap_qenter(kva, pages, npages);
- Remove 'struct vmmeter' from 'struct pcpu', leaving only global vmmeter in place. To do per-cpu stats, convert all fields that previously were maintained in the vmmeters that sit in pcpus to counter(9). - Since some vmmeter stats may be touched at very early stages of boot, before we have set up UMA and we can do counter_u64_alloc(), provide an early counter mechanism: o Leave one spare uint64_t in struct pcpu, named pc_early_dummy_counter. o Point counter(9) fields of vmmeter to pcpu[0].pc_early_dummy_counter, so that at early stages of boot, before counters are allocated we already point to a counter that can be safely written to. o For sparc64 that required a whole dummy pcpu[MAXCPU] array. Further related changes: - Don't include vmmeter.h into pcpu.h. - vm.stats.vm.v_swappgsout and vm.stats.vm.v_swappgsin changed to 64-bit, to match kernel representation. - struct vmmeter hidden under _KERNEL, and only vmstat(1) is an exclusion. This is based on benno@'s 4-year old patch: https://lists.freebsd.org/pipermail/freebsd-arch/2013-July/014471.html Reviewed by: kib, gallatin, marius, lidl Differential Revision: https://reviews.freebsd.org/D10156
2017-04-17 17:34:47 +00:00
VM_CNT_INC(v_vnodein);
VM_CNT_ADD(v_vnodepgsin, npages);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
Fix breakage caused by r292373 in ZFS/FUSE/NFS/SMBFS. With the new VOP_GETPAGES() KPI the "count" argument counts pages already, and doesn't need to be translated from bytes to pages. While here make it consistent that *rbehind and *rahead are updated only if we doesn't return error. Pointy hat to: glebius
2015-12-16 23:48:50 +00:00
count = npages << PAGE_SHIFT;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
iov.iov_base = (caddr_t) kva;
iov.iov_len = count;
uio.uio_iov = &iov;
uio.uio_iovcnt = 1;
uio.uio_offset = IDX_TO_OFF(pages[0]->pindex);
uio.uio_resid = count;
uio.uio_segflg = UIO_SYSSPACE;
uio.uio_rw = UIO_READ;
Catch up with KSE changes. Submitted by: Max Khon <fjoe@iclub.nsu.ru>
2001-12-02 08:56:58 +00:00
uio.uio_td = td;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
error = smb_read(smp->sm_share, np->n_fid, &uio, scred);
smbfs_free_scred(scred);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
pmap_qremove(kva, npages);
Allocate pager bufs from UMA instead of 80-ish mutex protected linked list. o In vm_pager_bufferinit() create pbuf_zone and start accounting on how many pbufs are we going to have set. In various subsystems that are going to utilize pbufs create private zones via call to pbuf_zsecond_create(). The latter calls uma_zsecond_create(), and sets a limit on created zone. After startup preallocate pbufs according to requirements of all pbuf zones. Subsystems that used to have a private limit with old allocator now have private pbuf zones: md(4), fusefs, NFS client, smbfs, VFS cluster, FFS, swap, vnode pager. The following subsystems use shared pbuf zone: cam(4), nvme(4), physio(9), aio(4). They should have their private limits, but changing that is out of scope of this commit. o Fetch tunable value of kern.nswbuf from init_param2() and while here move NSWBUF_MIN to opt_param.h and eliminate opt_swap.h, that was holding only this option. Default values aren't touched by this commit, but they probably should be reviewed wrt to modern hardware. This change removes a tight bottleneck from sendfile(2) operation, that uses pbufs in vnode pager. Other pagers also would benefit from faster allocation. Together with: gallatin Tested by: pho
2019-01-15 01:02:16 +00:00
uma_zfree(smbfs_pbuf_zone, bp);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
if (error && (uio.uio_resid == count)) {
printf("smbfs_getpages: error %d\n",error);
return VM_PAGER_ERROR;
}
size = count - uio.uio_resid;
A change to KPI of vm_pager_get_pages() and underlying VOP_GETPAGES(). o With new KPI consumers can request contiguous ranges of pages, and unlike before, all pages will be kept busied on return, like it was done before with the 'reqpage' only. Now the reqpage goes away. With new interface it is easier to implement code protected from race conditions. Such arrayed requests for now should be preceeded by a call to vm_pager_haspage() to make sure that request is possible. This could be improved later, making vm_pager_haspage() obsolete. Strenghtening the promises on the business of the array of pages allows us to remove such hacks as swp_pager_free_nrpage() and vm_pager_free_nonreq(). o New KPI accepts two integer pointers that may optionally point at values for read ahead and read behind, that a pager may do, if it can. These pages are completely owned by pager, and not controlled by the caller. This shifts the UFS-specific readahead logic from vm_fault.c, which should be file system agnostic, into vnode_pager.c. It also removes one VOP_BMAP() request per hard fault. Discussed with: kib, alc, jeff, scottl Sponsored by: Nginx, Inc. Sponsored by: Netflix
2015-12-16 21:30:45 +00:00
VM_OBJECT_WLOCK(object);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
for (i = 0, toff = 0; i < npages; i++, toff = nextoff) {
vm_page_t m;
nextoff = toff + PAGE_SIZE;
m = pages[i];
if (nextoff <= size) {
Track nfs's getpages() changes: Properly count v_vnodepgsin. Do not reread page if is already valid. Properly handle partially filled pages.
2002-04-23 14:30:43 +00:00
/*
* Read operation filled an entire page
*/
(4/6) Protect page valid with the busy lock. Atomics are used for page busy and valid state when the shared busy is held. The details of the locking protocol and valid and dirty synchronization are in the updated vm_page.h comments. Reviewed by: kib, markj Tested by: pho Sponsored by: Netflix, Intel Differential Revision: https://reviews.freebsd.org/D21594
2019-10-15 03:45:41 +00:00
vm_page_valid(m);
Eliminate gratuitous clearing of the page's dirty mask.
2009-05-12 05:49:02 +00:00
KASSERT(m->dirty == 0,
("smbfs_getpages: page %p is dirty", m));
Track nfs's getpages() changes: Properly count v_vnodepgsin. Do not reread page if is already valid. Properly handle partially filled pages.
2002-04-23 14:30:43 +00:00
} else if (size > toff) {
/*
* Read operation filled a partial page.
*/
(4/6) Protect page valid with the busy lock. Atomics are used for page busy and valid state when the shared busy is held. The details of the locking protocol and valid and dirty synchronization are in the updated vm_page.h comments. Reviewed by: kib, markj Tested by: pho Sponsored by: Netflix, Intel Differential Revision: https://reviews.freebsd.org/D21594
2019-10-15 03:45:41 +00:00
vm_page_invalid(m);
Rename vm_page_set_valid() to vm_page_set_valid_range(). The vm_page_set_valid() is the most reasonable name for the m->valid accessor. Reviewed by: attilio, alc
2011-11-30 17:39:00 +00:00
vm_page_set_valid_range(m, 0, size - toff);
Make *getpages()s' assertion on the state of each page's dirty bits stricter.
2009-05-28 18:11:09 +00:00
KASSERT(m->dirty == 0,
Eliminate unnecessary clearing of the page's dirty mask from various getpages functions. Eliminate a stale comment.
2009-05-15 04:33:35 +00:00
("smbfs_getpages: page %p is dirty", m));
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
} else {
Track nfs's getpages() changes: Properly count v_vnodepgsin. Do not reread page if is already valid. Properly handle partially filled pages.
2002-04-23 14:30:43 +00:00
/*
sys/fs: spelling fixes in comments. No functional change.
2016-04-29 20:51:24 +00:00
* Read operation was short. If no error occurred
Track nfs's getpages() changes: Properly count v_vnodepgsin. Do not reread page if is already valid. Properly handle partially filled pages.
2002-04-23 14:30:43 +00:00
* we may have hit a zero-fill section. We simply
* leave valid set to 0.
*/
;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
}
}
Fix breakage caused by r292373 in ZFS/FUSE/NFS/SMBFS. With the new VOP_GETPAGES() KPI the "count" argument counts pages already, and doesn't need to be translated from bytes to pages. While here make it consistent that *rbehind and *rahead are updated only if we doesn't return error. Pointy hat to: glebius
2015-12-16 23:48:50 +00:00
out:
Change VM_OBJECT_LOCK/UNLOCK() -> VM_OBJECT_WLOCK/WUNLOCK() to reflect the recent switch of the vm object lock to a rwlock. Reported by: attilio
2013-05-04 14:27:28 +00:00
VM_OBJECT_WUNLOCK(object);
Fix breakage caused by r292373 in ZFS/FUSE/NFS/SMBFS. With the new VOP_GETPAGES() KPI the "count" argument counts pages already, and doesn't need to be translated from bytes to pages. While here make it consistent that *rbehind and *rahead are updated only if we doesn't return error. Pointy hat to: glebius
2015-12-16 23:48:50 +00:00
if (ap->a_rbehind)
*ap->a_rbehind = 0;
if (ap->a_rahead)
*ap->a_rahead = 0;
return (VM_PAGER_OK);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
#endif /* SMBFS_RWGENERIC */
}
/*
* Vnode op for VM putpages.
* possible bug: all IO done in sync mode
* Note that vop_close always invalidate pages before close, so it's
* not necessary to open vnode.
*/
int
smbfs_putpages(ap)
struct vop_putpages_args /* {
struct vnode *a_vp;
vm_page_t *a_m;
int a_count;
int a_sync;
int *a_rtvals;
} */ *ap;
{
int error;
struct vnode *vp = ap->a_vp;
Catch up with KSE changes. Submitted by: Max Khon <fjoe@iclub.nsu.ru>
2001-12-02 08:56:58 +00:00
struct thread *td;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
struct ucred *cred;
#ifdef SMBFS_RWGENERIC
Catch up with KSE changes. Submitted by: Max Khon <fjoe@iclub.nsu.ru>
2001-12-02 08:56:58 +00:00
td = curthread; /* XXX */
Simple p_ucred -> td_ucred changes to start using the per-thread ucred reference.
2002-02-27 18:32:23 +00:00
cred = td->td_ucred; /* XXX */
Change the VOP_OPEN(), vn_open() vnode operation and d_fdopen() cdev operation argument from being file descriptor index into the pointer to struct file: part 2. Convert calls missed in the first big commit. Noted by: rwatson Pointy hat to: kib
2007-06-01 14:33:11 +00:00
VOP_OPEN(vp, FWRITE, cred, td, NULL);
Implement vop_std{get|put}pages() and add them to the default vop[]. Un-copy&paste all the VOP_{GET|PUT}PAGES() functions which do nothing but the default.
2001-05-01 08:34:45 +00:00
error = vop_stdputpages(ap);
Catch up with KSE changes. Submitted by: Max Khon <fjoe@iclub.nsu.ru>
2001-12-02 08:56:58 +00:00
VOP_CLOSE(vp, FWRITE, cred, td);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
return error;
#else
struct uio uio;
struct iovec iov;
vm_offset_t kva;
struct buf *bp;
int i, npages, count;
int *rtvals;
struct smbmount *smp;
struct smbnode *np;
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
struct smb_cred *scred;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
vm_page_t *pages;
Catch up with KSE changes. Submitted by: Max Khon <fjoe@iclub.nsu.ru>
2001-12-02 08:56:58 +00:00
td = curthread; /* XXX */
Simple p_ucred -> td_ucred changes to start using the per-thread ucred reference.
2002-02-27 18:32:23 +00:00
cred = td->td_ucred; /* XXX */
Change the VOP_OPEN(), vn_open() vnode operation and d_fdopen() cdev operation argument from being file descriptor index into the pointer to struct file: part 2. Convert calls missed in the first big commit. Noted by: rwatson Pointy hat to: kib
2007-06-01 14:33:11 +00:00
/* VOP_OPEN(vp, FWRITE, cred, td, NULL);*/
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
np = VTOSMB(vp);
smp = VFSTOSMBFS(vp->v_mount);
pages = ap->a_m;
count = ap->a_count;
rtvals = ap->a_rtvals;
npages = btoc(count);
for (i = 0; i < npages; i++) {
In the VOP_PUTPAGES() implementations, change the default error from VM_PAGER_AGAIN to VM_PAGER_ERROR for the uwritten pages. Return VM_PAGER_AGAIN for the partially written page. Always forward at least one page in the loop of vm_object_page_clean(). VM_PAGER_ERROR causes the page reactivation and does not clear the page dirty state, so the write is not lost. The change fixes an infinite loop in vm_object_page_clean() when the filesystem returns permanent errors for some page writes. Reported and tested by: gavin Reviewed by: alc, rmacklem MFC after: 1 week
2011-06-01 21:00:28 +00:00
rtvals[i] = VM_PAGER_ERROR;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
}
Allocate pager bufs from UMA instead of 80-ish mutex protected linked list. o In vm_pager_bufferinit() create pbuf_zone and start accounting on how many pbufs are we going to have set. In various subsystems that are going to utilize pbufs create private zones via call to pbuf_zsecond_create(). The latter calls uma_zsecond_create(), and sets a limit on created zone. After startup preallocate pbufs according to requirements of all pbuf zones. Subsystems that used to have a private limit with old allocator now have private pbuf zones: md(4), fusefs, NFS client, smbfs, VFS cluster, FFS, swap, vnode pager. The following subsystems use shared pbuf zone: cam(4), nvme(4), physio(9), aio(4). They should have their private limits, but changing that is out of scope of this commit. o Fetch tunable value of kern.nswbuf from init_param2() and while here move NSWBUF_MIN to opt_param.h and eliminate opt_swap.h, that was holding only this option. Default values aren't touched by this commit, but they probably should be reviewed wrt to modern hardware. This change removes a tight bottleneck from sendfile(2) operation, that uses pbufs in vnode pager. Other pagers also would benefit from faster allocation. Together with: gallatin Tested by: pho
2019-01-15 01:02:16 +00:00
bp = uma_zalloc(smbfs_pbuf_zone, M_WAITOK);
Get rid from extra #ifdefs.
2002-04-23 13:55:14 +00:00
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
kva = (vm_offset_t) bp->b_data;
pmap_qenter(kva, pages, npages);
- Remove 'struct vmmeter' from 'struct pcpu', leaving only global vmmeter in place. To do per-cpu stats, convert all fields that previously were maintained in the vmmeters that sit in pcpus to counter(9). - Since some vmmeter stats may be touched at very early stages of boot, before we have set up UMA and we can do counter_u64_alloc(), provide an early counter mechanism: o Leave one spare uint64_t in struct pcpu, named pc_early_dummy_counter. o Point counter(9) fields of vmmeter to pcpu[0].pc_early_dummy_counter, so that at early stages of boot, before counters are allocated we already point to a counter that can be safely written to. o For sparc64 that required a whole dummy pcpu[MAXCPU] array. Further related changes: - Don't include vmmeter.h into pcpu.h. - vm.stats.vm.v_swappgsout and vm.stats.vm.v_swappgsin changed to 64-bit, to match kernel representation. - struct vmmeter hidden under _KERNEL, and only vmstat(1) is an exclusion. This is based on benno@'s 4-year old patch: https://lists.freebsd.org/pipermail/freebsd-arch/2013-July/014471.html Reviewed by: kib, gallatin, marius, lidl Differential Revision: https://reviews.freebsd.org/D10156
2017-04-17 17:34:47 +00:00
VM_CNT_INC(v_vnodeout);
VM_CNT_ADD(v_vnodepgsout, count);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
iov.iov_base = (caddr_t) kva;
iov.iov_len = count;
uio.uio_iov = &iov;
uio.uio_iovcnt = 1;
uio.uio_offset = IDX_TO_OFF(pages[0]->pindex);
uio.uio_resid = count;
uio.uio_segflg = UIO_SYSSPACE;
uio.uio_rw = UIO_WRITE;
Catch up with KSE changes. Submitted by: Max Khon <fjoe@iclub.nsu.ru>
2001-12-02 08:56:58 +00:00
uio.uio_td = td;
Change the code to use %jd as printf() placeholder for uio_offset and cast to intmax_t. Suggested by: pjd Sponsored by: iXsystems inc.
2012-10-31 02:54:44 +00:00
SMBVDEBUG("ofs=%jd,resid=%zd\n", (intmax_t)uio.uio_offset,
uio.uio_resid);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
Fix panic due to page faults while in kernel mode, under conditions of VM pressure. The reason is that in some codepaths pointers to stack variables were passed from one thread to another. In collaboration with: pho Reported by: pho's stress2 suite Sponsored by: iXsystems inc.
2012-10-31 03:34:07 +00:00
scred = smbfs_malloc_scred();
smb_makescred(scred, td, cred);
error = smb_write(smp->sm_share, np->n_fid, &uio, scred);
smbfs_free_scred(scred);
Catch up with KSE changes. Submitted by: Max Khon <fjoe@iclub.nsu.ru>
2001-12-02 08:56:58 +00:00
/* VOP_CLOSE(vp, FWRITE, cred, td);*/
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
SMBVDEBUG("paged write done: %d\n", error);
pmap_qremove(kva, npages);
Get rid from extra #ifdefs.
2002-04-23 13:55:14 +00:00
Allocate pager bufs from UMA instead of 80-ish mutex protected linked list. o In vm_pager_bufferinit() create pbuf_zone and start accounting on how many pbufs are we going to have set. In various subsystems that are going to utilize pbufs create private zones via call to pbuf_zsecond_create(). The latter calls uma_zsecond_create(), and sets a limit on created zone. After startup preallocate pbufs according to requirements of all pbuf zones. Subsystems that used to have a private limit with old allocator now have private pbuf zones: md(4), fusefs, NFS client, smbfs, VFS cluster, FFS, swap, vnode pager. The following subsystems use shared pbuf zone: cam(4), nvme(4), physio(9), aio(4). They should have their private limits, but changing that is out of scope of this commit. o Fetch tunable value of kern.nswbuf from init_param2() and while here move NSWBUF_MIN to opt_param.h and eliminate opt_swap.h, that was holding only this option. Default values aren't touched by this commit, but they probably should be reviewed wrt to modern hardware. This change removes a tight bottleneck from sendfile(2) operation, that uses pbufs in vnode pager. Other pagers also would benefit from faster allocation. Together with: gallatin Tested by: pho
2019-01-15 01:02:16 +00:00
uma_zfree(smbfs_pbuf_zone, bp);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
Mark pages after EOF as clean after pageout. Suppose that a file on NFS has partially filled last page, and this page is dirty. NFS VOP_PAGEOUT() method only marks the the page clean up to the block of the last written byte, leaving other blocks dirty. Also any page which erronously exists in the vnode vm_object past EOF is also left marked as dirty. With the introduction of the buf-cache coherent pager, each pass of syncer over the object with such page results in creation of B_DELWRI buffer due to VOP_WRITE() call. This buffer is noted on next syncer pass, which results e.g. a visible manifestation of shutdown never finishing vnode sync. Note that before buf-cache coherency commit, a dirty page might left never synced to server if a partial writes occur. Fix this by clearing dirty bits after EOF. Only blocks of the partial page which are completely after EOF are marked clean, to avoid possible user data loss. Reported by: mav Reviewed by: alc, markj Tested by: mav, pho Sponsored by: The FreeBSD Foundation MFC after: 1 week Differential revision: https://reviews.freebsd.org/D11697
2017-07-26 20:07:05 +00:00
if (error == 0) {
vnode_pager_undirty_pages(pages, rtvals, count - uio.uio_resid,
npages * PAGE_SIZE, npages * PAGE_SIZE);
}
return (rtvals[0]);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
#endif /* SMBFS_RWGENERIC */
}
/*
* Flush and invalidate all dirty buffers. If another process is already
* doing the flush, just wait for completion.
*/
int
Eliminate unused and constant arguments to smbfs_vinvalbuf()
2005-01-14 08:52:55 +00:00
smbfs_vinvalbuf(struct vnode *vp, struct thread *td)
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
{
struct smbnode *np = VTOSMB(vp);
Eliminate unused and constant arguments to smbfs_vinvalbuf()
2005-01-14 08:52:55 +00:00
int error = 0;
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
vfs: introduce v_irflag and make v_type smaller The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time. v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED. Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
2019-12-08 21:30:04 +00:00
if (VN_IS_DOOMED(vp))
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
return 0;
- Replace v_flag with v_iflag and v_vflag - v_vflag is protected by the vnode lock and is used when synchronization with VOP calls is needed. - v_iflag is protected by interlock and is used for dealing with vnode management issues. These flags include X/O LOCK, FREE, DOOMED, etc. - All accesses to v_iflag and v_vflag have either been locked or marked with mp_fixme's. - Many ASSERT_VOP_LOCKED calls have been added where the locking was not clear. - Many functions in vfs_subr.c were restructured to provide for stronger locking. Idea stolen from: BSD/OS
2002-08-04 10:29:36 +00:00
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
while (np->n_flag & NFLUSHINPROG) {
np->n_flag |= NFLUSHWANT;
Eliminate unused and constant arguments to smbfs_vinvalbuf()
2005-01-14 08:52:55 +00:00
error = tsleep(&np->n_flag, PRIBIO + 2, "smfsvinv", 2 * hz);
- smb_td_intr takes a thread as an argument not a proc.
2003-04-01 09:24:12 +00:00
error = smb_td_intr(td);
Eliminate unused and constant arguments to smbfs_vinvalbuf()
2005-01-14 08:52:55 +00:00
if (error == EINTR)
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
return EINTR;
}
np->n_flag |= NFLUSHINPROG;
Do not set B_NOCACHE on buffers when releasing them in flushbuflist(). If B_NOCACHE is set the pages of vm backed buffers will be invalidated. However clean buffers can be backed by dirty VM pages so invalidating them can lead to data loss. Add support for flush dirty page in the data invalidation function of some network file systems. This fixes data losses during vnode recycling (and other code paths using invalbuf(*,V_SAVE,*,*)) for data written using an mmaped file. Collaborative effort by: jhb@,mohans@,peter@,ps@,ups@ Reviewed by: tegge@ MFC after: 7 days
2006-05-25 01:00:35 +00:00
Call vm_object_page_clean() with the object lock held. Submitted by: kensmith@ Reviewed by: mohans@ MFC after: 6 days
2006-05-25 17:16:11 +00:00
if (vp->v_bufobj.bo_object != NULL) {
Change VM_OBJECT_LOCK/UNLOCK() -> VM_OBJECT_WLOCK/WUNLOCK() to reflect the recent switch of the vm object lock to a rwlock. Reported by: attilio
2013-05-04 14:27:28 +00:00
VM_OBJECT_WLOCK(vp->v_bufobj.bo_object);
Do not set B_NOCACHE on buffers when releasing them in flushbuflist(). If B_NOCACHE is set the pages of vm backed buffers will be invalidated. However clean buffers can be backed by dirty VM pages so invalidating them can lead to data loss. Add support for flush dirty page in the data invalidation function of some network file systems. This fixes data losses during vnode recycling (and other code paths using invalbuf(*,V_SAVE,*,*)) for data written using an mmaped file. Collaborative effort by: jhb@,mohans@,peter@,ps@,ups@ Reviewed by: tegge@ MFC after: 7 days
2006-05-25 01:00:35 +00:00
vm_object_page_clean(vp->v_bufobj.bo_object, 0, 0, OBJPC_SYNC);
Change VM_OBJECT_LOCK/UNLOCK() -> VM_OBJECT_WLOCK/WUNLOCK() to reflect the recent switch of the vm object lock to a rwlock. Reported by: attilio
2013-05-04 14:27:28 +00:00
VM_OBJECT_WUNLOCK(vp->v_bufobj.bo_object);
Call vm_object_page_clean() with the object lock held. Submitted by: kensmith@ Reviewed by: mohans@ MFC after: 6 days
2006-05-25 17:16:11 +00:00
}
Do not set B_NOCACHE on buffers when releasing them in flushbuflist(). If B_NOCACHE is set the pages of vm backed buffers will be invalidated. However clean buffers can be backed by dirty VM pages so invalidating them can lead to data loss. Add support for flush dirty page in the data invalidation function of some network file systems. This fixes data losses during vnode recycling (and other code paths using invalbuf(*,V_SAVE,*,*)) for data written using an mmaped file. Collaborative effort by: jhb@,mohans@,peter@,ps@,ups@ Reviewed by: tegge@ MFC after: 7 days
2006-05-25 01:00:35 +00:00
Remove the struct thread unuseful argument from bufobj interface. In particular following functions KPI results modified: - bufobj_invalbuf() - bufsync() and BO_SYNC() "virtual method" of the buffer objects set. Main consumers of bufobj functions are affected by this change too and, in particular, functions which changed their KPI are: - vinvalbuf() - g_vfs_close() Due to the KPI breakage, __FreeBSD_version will be bumped in a later commit. As a side note, please consider just temporary the 'curthread' argument passing to VOP_SYNC() (in bufsync()) as it will be axed out ASAP Reviewed by: kib Tested by: Giovanni Trematerra <giovanni dot trematerra at gmail dot com>
2008-10-10 21:23:50 +00:00
error = vinvalbuf(vp, V_SAVE, PCATCH, 0);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
while (error) {
Eliminate unused and constant arguments to smbfs_vinvalbuf()
2005-01-14 08:52:55 +00:00
if (error == ERESTART || error == EINTR) {
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
np->n_flag &= ~NFLUSHINPROG;
if (np->n_flag & NFLUSHWANT) {
np->n_flag &= ~NFLUSHWANT;
More low-hanging fruit: kill caddr_t in calls to wakeup(9) / [mt]sleep(9).
2003-03-02 16:54:40 +00:00
wakeup(&np->n_flag);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
}
return EINTR;
}
Remove the struct thread unuseful argument from bufobj interface. In particular following functions KPI results modified: - bufobj_invalbuf() - bufsync() and BO_SYNC() "virtual method" of the buffer objects set. Main consumers of bufobj functions are affected by this change too and, in particular, functions which changed their KPI are: - vinvalbuf() - g_vfs_close() Due to the KPI breakage, __FreeBSD_version will be bumped in a later commit. As a side note, please consider just temporary the 'curthread' argument passing to VOP_SYNC() (in bufsync()) as it will be axed out ASAP Reviewed by: kib Tested by: Giovanni Trematerra <giovanni dot trematerra at gmail dot com>
2008-10-10 21:23:50 +00:00
error = vinvalbuf(vp, V_SAVE, PCATCH, 0);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
}
np->n_flag &= ~(NMODIFIED | NFLUSHINPROG);
if (np->n_flag & NFLUSHWANT) {
np->n_flag &= ~NFLUSHWANT;
More low-hanging fruit: kill caddr_t in calls to wakeup(9) / [mt]sleep(9).
2003-03-02 16:54:40 +00:00
wakeup(&np->n_flag);
Import kernel part of SMB/CIFS requester. Add smbfs(CIFS) filesystem. Userland part will be in the ports tree for a while. Obtained from: smbfs-1.3.7-dev package.
2001-04-10 07:59:06 +00:00
}
return (error);
}
Reference in New Issue Copy Permalink