1994-05-24 10:09:53 +00:00
|
|
|
/*
|
|
|
|
|
* Copyright (c) 1982, 1986, 1989, 1993
|
|
|
|
|
* The Regents of the University of California. All rights reserved.
|
|
|
|
|
* (c) UNIX System Laboratories, Inc.
|
|
|
|
|
* All or some portions of this file are derived from material licensed
|
|
|
|
|
* to the University of California by American Telephone and Telegraph
|
|
|
|
|
* Co. or Unix System Laboratories, Inc. and are reproduced herein with
|
|
|
|
|
* the permission of UNIX System Laboratories, Inc.
|
|
|
|
|
*
|
|
|
|
|
* 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. All advertising materials mentioning features or use of this software
|
|
|
|
|
* must display the following acknowledgement:
|
|
|
|
|
* This product includes software developed by the University of
|
|
|
|
|
* California, Berkeley and its contributors.
|
|
|
|
|
* 4. Neither the name of the University nor the names of its contributors
|
|
|
|
|
* may be used to endorse or promote products derived from this software
|
|
|
|
|
* without specific prior written permission.
|
|
|
|
|
*
|
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
|
|
|
|
|
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
|
|
|
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
|
|
|
|
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
|
|
|
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
|
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
|
|
|
|
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
|
|
|
|
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
|
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
|
|
|
|
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
|
|
|
|
* SUCH DAMAGE.
|
|
|
|
|
*
|
|
|
|
|
* @(#)vfs_vnops.c 8.2 (Berkeley) 1/21/94
|
1999-08-28 01:08:13 +00:00
|
|
|
* $FreeBSD$
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
|
|
|
|
|
|
|
|
|
#include <sys/param.h>
|
|
|
|
|
#include <sys/systm.h>
|
1997-03-23 03:37:54 +00:00
|
|
|
#include <sys/fcntl.h>
|
1994-05-24 10:09:53 +00:00
|
|
|
#include <sys/file.h>
|
|
|
|
|
#include <sys/stat.h>
|
|
|
|
|
#include <sys/proc.h>
|
|
|
|
|
#include <sys/mount.h>
|
2000-10-20 07:58:15 +00:00
|
|
|
#include <sys/mutex.h>
|
1994-05-24 10:09:53 +00:00
|
|
|
#include <sys/namei.h>
|
|
|
|
|
#include <sys/vnode.h>
|
2000-05-05 09:59:14 +00:00
|
|
|
#include <sys/bio.h>
|
1999-08-04 18:53:50 +00:00
|
|
|
#include <sys/buf.h>
|
1997-03-24 11:52:29 +00:00
|
|
|
#include <sys/filio.h>
|
|
|
|
|
#include <sys/ttycom.h>
|
1999-08-13 11:22:48 +00:00
|
|
|
#include <sys/conf.h>
|
1994-05-24 10:09:53 +00:00
|
|
|
|
1995-12-17 21:23:44 +00:00
|
|
|
static int vn_closefile __P((struct file *fp, struct proc *p));
|
1998-06-07 17:13:14 +00:00
|
|
|
static int vn_ioctl __P((struct file *fp, u_long com, caddr_t data,
|
1995-12-17 21:23:44 +00:00
|
|
|
struct proc *p));
|
|
|
|
|
static int vn_read __P((struct file *fp, struct uio *uio,
|
This is what was "fdfix2.patch," a fix for fd sharing. It's pretty
far-reaching in fd-land, so you'll want to consult the code for
changes. The biggest change is that now, you don't use
fp->f_ops->fo_foo(fp, bar)
but instead
fo_foo(fp, bar),
which increments and decrements the fp refcount upon entry and exit.
Two new calls, fhold() and fdrop(), are provided. Each does what it
seems like it should, and if fdrop() brings the refcount to zero, the
fd is freed as well.
Thanks to peter ("to hell with it, it looks ok to me.") for his review.
Thanks to msmith for keeping me from putting locks everywhere :)
Reviewed by: peter
1999-09-19 17:00:25 +00:00
|
|
|
struct ucred *cred, int flags, struct proc *p));
|
1997-09-14 02:51:16 +00:00
|
|
|
static int vn_poll __P((struct file *fp, int events, struct ucred *cred,
|
|
|
|
|
struct proc *p));
|
2001-02-15 16:34:11 +00:00
|
|
|
static int vn_kqfilter __P((struct file *fp, struct knote *kn));
|
1999-11-08 03:32:15 +00:00
|
|
|
static int vn_statfile __P((struct file *fp, struct stat *sb, struct proc *p));
|
1995-12-17 21:23:44 +00:00
|
|
|
static int vn_write __P((struct file *fp, struct uio *uio,
|
This is what was "fdfix2.patch," a fix for fd sharing. It's pretty
far-reaching in fd-land, so you'll want to consult the code for
changes. The biggest change is that now, you don't use
fp->f_ops->fo_foo(fp, bar)
but instead
fo_foo(fp, bar),
which increments and decrements the fp refcount upon entry and exit.
Two new calls, fhold() and fdrop(), are provided. Each does what it
seems like it should, and if fdrop() brings the refcount to zero, the
fd is freed as well.
Thanks to peter ("to hell with it, it looks ok to me.") for his review.
Thanks to msmith for keeping me from putting locks everywhere :)
Reviewed by: peter
1999-09-19 17:00:25 +00:00
|
|
|
struct ucred *cred, int flags, struct proc *p));
|
1995-12-17 21:23:44 +00:00
|
|
|
|
2001-02-15 16:34:11 +00:00
|
|
|
struct fileops vnops = {
|
|
|
|
|
vn_read, vn_write, vn_ioctl, vn_poll, vn_kqfilter,
|
|
|
|
|
vn_statfile, vn_closefile
|
2000-04-16 18:53:38 +00:00
|
|
|
};
|
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
|
|
|
|
* Common code for vnode open operations.
|
|
|
|
|
* Check permissions, and call the VOP_OPEN or VOP_CREATE routine.
|
1999-12-15 23:02:35 +00:00
|
|
|
*
|
2000-05-12 16:06:49 +00:00
|
|
|
* Note that this does NOT free nameidata for the successful case,
|
1999-12-15 23:02:35 +00:00
|
|
|
* due to the NDINIT being done elsewhere.
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
1994-05-25 09:21:21 +00:00
|
|
|
int
|
2000-07-04 03:34:11 +00:00
|
|
|
vn_open(ndp, flagp, cmode)
|
1994-05-24 10:09:53 +00:00
|
|
|
register struct nameidata *ndp;
|
2000-07-04 03:34:11 +00:00
|
|
|
int *flagp, cmode;
|
1994-05-24 10:09:53 +00:00
|
|
|
{
|
2000-07-04 03:34:11 +00:00
|
|
|
struct vnode *vp;
|
2000-07-11 22:07:57 +00:00
|
|
|
struct mount *mp;
|
2000-07-04 03:34:11 +00:00
|
|
|
struct proc *p = ndp->ni_cnd.cn_proc;
|
|
|
|
|
struct ucred *cred = p->p_ucred;
|
1994-05-24 10:09:53 +00:00
|
|
|
struct vattr vat;
|
|
|
|
|
struct vattr *vap = &vat;
|
2000-07-04 03:34:11 +00:00
|
|
|
int mode, fmode, error;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
2000-07-11 22:07:57 +00:00
|
|
|
restart:
|
2000-07-04 03:34:11 +00:00
|
|
|
fmode = *flagp;
|
1994-05-24 10:09:53 +00:00
|
|
|
if (fmode & O_CREAT) {
|
|
|
|
|
ndp->ni_cnd.cn_nameiop = CREATE;
|
|
|
|
|
ndp->ni_cnd.cn_flags = LOCKPARENT | LOCKLEAF;
|
1998-04-06 18:25:21 +00:00
|
|
|
if ((fmode & O_EXCL) == 0 && (fmode & O_NOFOLLOW) == 0)
|
1994-05-24 10:09:53 +00:00
|
|
|
ndp->ni_cnd.cn_flags |= FOLLOW;
|
2000-01-10 00:08:53 +00:00
|
|
|
bwillwrite();
|
2000-07-04 03:34:11 +00:00
|
|
|
if ((error = namei(ndp)) != 0)
|
1994-05-24 10:09:53 +00:00
|
|
|
return (error);
|
|
|
|
|
if (ndp->ni_vp == NULL) {
|
|
|
|
|
VATTR_NULL(vap);
|
|
|
|
|
vap->va_type = VREG;
|
|
|
|
|
vap->va_mode = cmode;
|
1997-02-10 02:22:35 +00:00
|
|
|
if (fmode & O_EXCL)
|
|
|
|
|
vap->va_vaflags |= VA_EXCLUSIVE;
|
2000-07-11 22:07:57 +00:00
|
|
|
if (vn_start_write(ndp->ni_dvp, &mp, V_NOWAIT) != 0) {
|
|
|
|
|
NDFREE(ndp, NDF_ONLY_PNBUF);
|
|
|
|
|
vput(ndp->ni_dvp);
|
|
|
|
|
if ((error = vn_start_write(NULL, &mp,
|
|
|
|
|
V_XSLEEP | PCATCH)) != 0)
|
|
|
|
|
return (error);
|
|
|
|
|
goto restart;
|
|
|
|
|
}
|
1997-02-10 02:22:35 +00:00
|
|
|
VOP_LEASE(ndp->ni_dvp, p, cred, LEASE_WRITE);
|
1998-05-07 04:58:58 +00:00
|
|
|
error = VOP_CREATE(ndp->ni_dvp, &ndp->ni_vp,
|
|
|
|
|
&ndp->ni_cnd, vap);
|
2000-07-04 03:34:11 +00:00
|
|
|
vput(ndp->ni_dvp);
|
2000-07-11 22:07:57 +00:00
|
|
|
vn_finished_write(mp);
|
1999-12-15 23:02:35 +00:00
|
|
|
if (error) {
|
|
|
|
|
NDFREE(ndp, NDF_ONLY_PNBUF);
|
1994-05-24 10:09:53 +00:00
|
|
|
return (error);
|
1999-12-15 23:02:35 +00:00
|
|
|
}
|
1997-04-04 17:46:21 +00:00
|
|
|
ASSERT_VOP_UNLOCKED(ndp->ni_dvp, "create");
|
|
|
|
|
ASSERT_VOP_LOCKED(ndp->ni_vp, "create");
|
1994-05-24 10:09:53 +00:00
|
|
|
fmode &= ~O_TRUNC;
|
|
|
|
|
vp = ndp->ni_vp;
|
|
|
|
|
} else {
|
|
|
|
|
if (ndp->ni_dvp == ndp->ni_vp)
|
|
|
|
|
vrele(ndp->ni_dvp);
|
|
|
|
|
else
|
|
|
|
|
vput(ndp->ni_dvp);
|
|
|
|
|
ndp->ni_dvp = NULL;
|
|
|
|
|
vp = ndp->ni_vp;
|
|
|
|
|
if (fmode & O_EXCL) {
|
|
|
|
|
error = EEXIST;
|
|
|
|
|
goto bad;
|
|
|
|
|
}
|
|
|
|
|
fmode &= ~O_CREAT;
|
|
|
|
|
}
|
|
|
|
|
} else {
|
|
|
|
|
ndp->ni_cnd.cn_nameiop = LOOKUP;
|
1998-04-06 17:38:43 +00:00
|
|
|
ndp->ni_cnd.cn_flags =
|
|
|
|
|
((fmode & O_NOFOLLOW) ? NOFOLLOW : FOLLOW) | LOCKLEAF;
|
2000-07-04 03:34:11 +00:00
|
|
|
if ((error = namei(ndp)) != 0)
|
1994-05-24 10:09:53 +00:00
|
|
|
return (error);
|
|
|
|
|
vp = ndp->ni_vp;
|
|
|
|
|
}
|
1998-04-06 18:43:28 +00:00
|
|
|
if (vp->v_type == VLNK) {
|
1998-04-06 19:32:37 +00:00
|
|
|
error = EMLINK;
|
1998-04-06 18:43:28 +00:00
|
|
|
goto bad;
|
|
|
|
|
}
|
1998-04-06 19:32:37 +00:00
|
|
|
if (vp->v_type == VSOCK) {
|
1994-05-24 10:09:53 +00:00
|
|
|
error = EOPNOTSUPP;
|
|
|
|
|
goto bad;
|
|
|
|
|
}
|
|
|
|
|
if ((fmode & O_CREAT) == 0) {
|
1998-11-02 02:36:16 +00:00
|
|
|
mode = 0;
|
1994-05-24 10:09:53 +00:00
|
|
|
if (fmode & (FWRITE | O_TRUNC)) {
|
|
|
|
|
if (vp->v_type == VDIR) {
|
|
|
|
|
error = EISDIR;
|
|
|
|
|
goto bad;
|
|
|
|
|
}
|
1994-10-02 17:35:40 +00:00
|
|
|
error = vn_writechk(vp);
|
1995-05-10 18:59:11 +00:00
|
|
|
if (error)
|
1994-10-02 17:35:40 +00:00
|
|
|
goto bad;
|
1998-11-02 02:36:16 +00:00
|
|
|
mode |= VWRITE;
|
|
|
|
|
}
|
|
|
|
|
if (fmode & FREAD)
|
|
|
|
|
mode |= VREAD;
|
|
|
|
|
if (mode) {
|
|
|
|
|
error = VOP_ACCESS(vp, mode, cred, p);
|
1995-05-10 18:59:11 +00:00
|
|
|
if (error)
|
1994-05-24 10:09:53 +00:00
|
|
|
goto bad;
|
|
|
|
|
}
|
|
|
|
|
}
|
2000-07-04 03:34:11 +00:00
|
|
|
if ((error = VOP_OPEN(vp, fmode, cred, p)) != 0)
|
1994-05-24 10:09:53 +00:00
|
|
|
goto bad;
|
1994-10-05 09:48:45 +00:00
|
|
|
/*
|
1996-08-21 21:56:23 +00:00
|
|
|
* Make sure that a VM object is created for VMIO support.
|
1994-10-05 09:48:45 +00:00
|
|
|
*/
|
1999-07-26 06:25:53 +00:00
|
|
|
if (vn_canvmio(vp) == TRUE) {
|
1999-01-05 18:50:03 +00:00
|
|
|
if ((error = vfs_object_create(vp, p, cred)) != 0)
|
1995-12-11 04:58:34 +00:00
|
|
|
goto bad;
|
1994-10-05 09:48:45 +00:00
|
|
|
}
|
1996-08-21 21:56:23 +00:00
|
|
|
|
NOTE: libkvm, w, ps, 'top', and any other utility which depends on struct
proc or any VM system structure will have to be rebuilt!!!
Much needed overhaul of the VM system. Included in this first round of
changes:
1) Improved pager interfaces: init, alloc, dealloc, getpages, putpages,
haspage, and sync operations are supported. The haspage interface now
provides information about clusterability. All pager routines now take
struct vm_object's instead of "pagers".
2) Improved data structures. In the previous paradigm, there is constant
confusion caused by pagers being both a data structure ("allocate a
pager") and a collection of routines. The idea of a pager structure has
escentially been eliminated. Objects now have types, and this type is
used to index the appropriate pager. In most cases, items in the pager
structure were duplicated in the object data structure and thus were
unnecessary. In the few cases that remained, a un_pager structure union
was created in the object to contain these items.
3) Because of the cleanup of #1 & #2, a lot of unnecessary layering can now
be removed. For instance, vm_object_enter(), vm_object_lookup(),
vm_object_remove(), and the associated object hash list were some of the
things that were removed.
4) simple_lock's removed. Discussion with several people reveals that the
SMP locking primitives used in the VM system aren't likely the mechanism
that we'll be adopting. Even if it were, the locking that was in the code
was very inadequate and would have to be mostly re-done anyway. The
locking in a uni-processor kernel was a no-op but went a long way toward
making the code difficult to read and debug.
5) Places that attempted to kludge-up the fact that we don't have kernel
thread support have been fixed to reflect the reality that we are really
dealing with processes, not threads. The VM system didn't have complete
thread support, so the comments and mis-named routines were just wrong.
We now use tsleep and wakeup directly in the lock routines, for instance.
6) Where appropriate, the pagers have been improved, especially in the
pager_alloc routines. Most of the pager_allocs have been rewritten and
are now faster and easier to maintain.
7) The pagedaemon pageout clustering algorithm has been rewritten and
now tries harder to output an even number of pages before and after
the requested page. This is sort of the reverse of the ideal pagein
algorithm and should provide better overall performance.
8) Unnecessary (incorrect) casts to caddr_t in calls to tsleep & wakeup
have been removed. Some other unnecessary casts have also been removed.
9) Some almost useless debugging code removed.
10) Terminology of shadow objects vs. backing objects straightened out.
The fact that the vm_object data structure escentially had this
backwards really confused things. The use of "shadow" and "backing
object" throughout the code is now internally consistent and correct
in the Mach terminology.
11) Several minor bug fixes, including one in the vm daemon that caused
0 RSS objects to not get purged as intended.
12) A "default pager" has now been created which cleans up the transition
of objects to the "swap" type. The previous checks throughout the code
for swp->pg_data != NULL were really ugly. This change also provides
the rudiments for future backing of "anonymous" memory by something
other than the swap pager (via the vnode pager, for example), and it
allows the decision about which of these pagers to use to be made
dynamically (although will need some additional decision code to do
this, of course).
13) (dyson) MAP_COPY has been deprecated and the corresponding "copy
object" code has been removed. MAP_COPY was undocumented and non-
standard. It was furthermore broken in several ways which caused its
behavior to degrade to MAP_PRIVATE. Binaries that use MAP_COPY will
continue to work correctly, but via the slightly different semantics
of MAP_PRIVATE.
14) (dyson) Sharing maps have been removed. It's marginal usefulness in a
threads design can be worked around in other ways. Both #12 and #13
were done to simplify the code and improve readability and maintain-
ability. (As were most all of these changes)
TODO:
1) Rewrite most of the vnode pager to use VOP_GETPAGES/PUTPAGES. Doing
this will reduce the vnode pager to a mere fraction of its current size.
2) Rewrite vm_fault and the swap/vnode pagers to use the clustering
information provided by the new haspage pager interface. This will
substantially reduce the overhead by eliminating a large number of
VOP_BMAP() calls. The VOP_BMAP() filesystem interface should be
improved to provide both a "behind" and "ahead" indication of
contiguousness.
3) Implement the extended features of pager_haspage in swap_pager_haspage().
It currently just says 0 pages ahead/behind.
4) Re-implement the swap device (swstrategy) in a more elegant way, perhaps
via a much more general mechanism that could also be used for disk
striping of regular filesystems.
5) Do something to improve the architecture of vm_object_collapse(). The
fact that it makes calls into the swap pager and knows too much about
how the swap pager operates really bothers me. It also doesn't allow
for collapsing of non-swap pager objects ("unnamed" objects backed by
other pagers).
1995-07-13 08:48:48 +00:00
|
|
|
if (fmode & FWRITE)
|
|
|
|
|
vp->v_writecount++;
|
2000-07-04 03:34:11 +00:00
|
|
|
*flagp = fmode;
|
1994-05-24 10:09:53 +00:00
|
|
|
return (0);
|
|
|
|
|
bad:
|
1999-12-15 23:02:35 +00:00
|
|
|
NDFREE(ndp, NDF_ONLY_PNBUF);
|
1994-05-24 10:09:53 +00:00
|
|
|
vput(vp);
|
2000-07-04 03:34:11 +00:00
|
|
|
*flagp = fmode;
|
1994-05-24 10:09:53 +00:00
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Check for write permissions on the specified vnode.
|
1997-02-10 02:22:35 +00:00
|
|
|
* Prototype text segments cannot be written.
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
1994-05-25 09:21:21 +00:00
|
|
|
int
|
1994-05-24 10:09:53 +00:00
|
|
|
vn_writechk(vp)
|
|
|
|
|
register struct vnode *vp;
|
|
|
|
|
{
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* If there's shared text associated with
|
|
|
|
|
* the vnode, try to free it up once. If
|
|
|
|
|
* we fail, we can't allow writing.
|
|
|
|
|
*/
|
1995-03-19 12:08:03 +00:00
|
|
|
if (vp->v_flag & VTEXT)
|
1994-05-24 10:09:53 +00:00
|
|
|
return (ETXTBSY);
|
|
|
|
|
return (0);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Vnode close call
|
|
|
|
|
*/
|
1994-05-25 09:21:21 +00:00
|
|
|
int
|
1994-05-24 10:09:53 +00:00
|
|
|
vn_close(vp, flags, cred, p)
|
|
|
|
|
register struct vnode *vp;
|
|
|
|
|
int flags;
|
|
|
|
|
struct ucred *cred;
|
|
|
|
|
struct proc *p;
|
|
|
|
|
{
|
|
|
|
|
int error;
|
|
|
|
|
|
|
|
|
|
if (flags & FWRITE)
|
|
|
|
|
vp->v_writecount--;
|
|
|
|
|
error = VOP_CLOSE(vp, flags, cred, p);
|
1996-08-21 21:56:23 +00:00
|
|
|
vrele(vp);
|
1994-05-24 10:09:53 +00:00
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
|
2000-04-02 00:55:28 +00:00
|
|
|
static __inline
|
|
|
|
|
int
|
|
|
|
|
sequential_heuristic(struct uio *uio, struct file *fp)
|
|
|
|
|
{
|
|
|
|
|
/*
|
|
|
|
|
* Sequential heuristic - detect sequential operation
|
|
|
|
|
*/
|
|
|
|
|
if ((uio->uio_offset == 0 && fp->f_seqcount > 0) ||
|
|
|
|
|
uio->uio_offset == fp->f_nextoff) {
|
|
|
|
|
/*
|
|
|
|
|
* XXX we assume that the filesystem block size is
|
|
|
|
|
* the default. Not true, but still gives us a pretty
|
|
|
|
|
* good indicator of how sequential the read operations
|
|
|
|
|
* are.
|
|
|
|
|
*/
|
|
|
|
|
fp->f_seqcount += (uio->uio_resid + BKVASIZE - 1) / BKVASIZE;
|
|
|
|
|
if (fp->f_seqcount >= 127)
|
|
|
|
|
fp->f_seqcount = 127;
|
|
|
|
|
return(fp->f_seqcount << 16);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Not sequential, quick draw-down of seqcount
|
|
|
|
|
*/
|
|
|
|
|
if (fp->f_seqcount > 1)
|
|
|
|
|
fp->f_seqcount = 1;
|
|
|
|
|
else
|
|
|
|
|
fp->f_seqcount = 0;
|
|
|
|
|
return(0);
|
|
|
|
|
}
|
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
|
|
|
|
* Package up an I/O request on a vnode into a uio and do it.
|
|
|
|
|
*/
|
1994-05-25 09:21:21 +00:00
|
|
|
int
|
1994-05-24 10:09:53 +00:00
|
|
|
vn_rdwr(rw, vp, base, len, offset, segflg, ioflg, cred, aresid, p)
|
|
|
|
|
enum uio_rw rw;
|
|
|
|
|
struct vnode *vp;
|
|
|
|
|
caddr_t base;
|
|
|
|
|
int len;
|
|
|
|
|
off_t offset;
|
|
|
|
|
enum uio_seg segflg;
|
|
|
|
|
int ioflg;
|
|
|
|
|
struct ucred *cred;
|
|
|
|
|
int *aresid;
|
|
|
|
|
struct proc *p;
|
|
|
|
|
{
|
|
|
|
|
struct uio auio;
|
|
|
|
|
struct iovec aiov;
|
2000-07-11 22:07:57 +00:00
|
|
|
struct mount *mp;
|
1994-05-24 10:09:53 +00:00
|
|
|
int error;
|
|
|
|
|
|
2000-07-11 22:07:57 +00:00
|
|
|
if ((ioflg & IO_NODELOCKED) == 0) {
|
|
|
|
|
mp = NULL;
|
|
|
|
|
if (rw == UIO_WRITE &&
|
2000-11-02 21:14:13 +00:00
|
|
|
vp->v_type != VCHR &&
|
2000-07-11 22:07:57 +00:00
|
|
|
(error = vn_start_write(vp, &mp, V_WAIT | PCATCH)) != 0)
|
|
|
|
|
return (error);
|
1997-02-10 02:22:35 +00:00
|
|
|
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, p);
|
2000-07-11 22:07:57 +00:00
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
auio.uio_iov = &aiov;
|
|
|
|
|
auio.uio_iovcnt = 1;
|
|
|
|
|
aiov.iov_base = base;
|
|
|
|
|
aiov.iov_len = len;
|
|
|
|
|
auio.uio_resid = len;
|
|
|
|
|
auio.uio_offset = offset;
|
|
|
|
|
auio.uio_segflg = segflg;
|
|
|
|
|
auio.uio_rw = rw;
|
|
|
|
|
auio.uio_procp = p;
|
|
|
|
|
if (rw == UIO_READ) {
|
|
|
|
|
error = VOP_READ(vp, &auio, ioflg, cred);
|
|
|
|
|
} else {
|
|
|
|
|
error = VOP_WRITE(vp, &auio, ioflg, cred);
|
|
|
|
|
}
|
|
|
|
|
if (aresid)
|
|
|
|
|
*aresid = auio.uio_resid;
|
|
|
|
|
else
|
|
|
|
|
if (auio.uio_resid && error == 0)
|
|
|
|
|
error = EIO;
|
2000-07-11 22:07:57 +00:00
|
|
|
if ((ioflg & IO_NODELOCKED) == 0) {
|
|
|
|
|
vn_finished_write(mp);
|
1997-02-10 02:22:35 +00:00
|
|
|
VOP_UNLOCK(vp, 0, p);
|
2000-07-11 22:07:57 +00:00
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* File table vnode read routine.
|
|
|
|
|
*/
|
1995-12-17 21:23:44 +00:00
|
|
|
static int
|
This is what was "fdfix2.patch," a fix for fd sharing. It's pretty
far-reaching in fd-land, so you'll want to consult the code for
changes. The biggest change is that now, you don't use
fp->f_ops->fo_foo(fp, bar)
but instead
fo_foo(fp, bar),
which increments and decrements the fp refcount upon entry and exit.
Two new calls, fhold() and fdrop(), are provided. Each does what it
seems like it should, and if fdrop() brings the refcount to zero, the
fd is freed as well.
Thanks to peter ("to hell with it, it looks ok to me.") for his review.
Thanks to msmith for keeping me from putting locks everywhere :)
Reviewed by: peter
1999-09-19 17:00:25 +00:00
|
|
|
vn_read(fp, uio, cred, flags, p)
|
1994-05-24 10:09:53 +00:00
|
|
|
struct file *fp;
|
|
|
|
|
struct uio *uio;
|
|
|
|
|
struct ucred *cred;
|
This is what was "fdfix2.patch," a fix for fd sharing. It's pretty
far-reaching in fd-land, so you'll want to consult the code for
changes. The biggest change is that now, you don't use
fp->f_ops->fo_foo(fp, bar)
but instead
fo_foo(fp, bar),
which increments and decrements the fp refcount upon entry and exit.
Two new calls, fhold() and fdrop(), are provided. Each does what it
seems like it should, and if fdrop() brings the refcount to zero, the
fd is freed as well.
Thanks to peter ("to hell with it, it looks ok to me.") for his review.
Thanks to msmith for keeping me from putting locks everywhere :)
Reviewed by: peter
1999-09-19 17:00:25 +00:00
|
|
|
struct proc *p;
|
1999-04-04 21:41:28 +00:00
|
|
|
int flags;
|
1994-05-24 10:09:53 +00:00
|
|
|
{
|
This is what was "fdfix2.patch," a fix for fd sharing. It's pretty
far-reaching in fd-land, so you'll want to consult the code for
changes. The biggest change is that now, you don't use
fp->f_ops->fo_foo(fp, bar)
but instead
fo_foo(fp, bar),
which increments and decrements the fp refcount upon entry and exit.
Two new calls, fhold() and fdrop(), are provided. Each does what it
seems like it should, and if fdrop() brings the refcount to zero, the
fd is freed as well.
Thanks to peter ("to hell with it, it looks ok to me.") for his review.
Thanks to msmith for keeping me from putting locks everywhere :)
Reviewed by: peter
1999-09-19 17:00:25 +00:00
|
|
|
struct vnode *vp;
|
1999-04-21 05:56:45 +00:00
|
|
|
int error, ioflag;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
This is what was "fdfix2.patch," a fix for fd sharing. It's pretty
far-reaching in fd-land, so you'll want to consult the code for
changes. The biggest change is that now, you don't use
fp->f_ops->fo_foo(fp, bar)
but instead
fo_foo(fp, bar),
which increments and decrements the fp refcount upon entry and exit.
Two new calls, fhold() and fdrop(), are provided. Each does what it
seems like it should, and if fdrop() brings the refcount to zero, the
fd is freed as well.
Thanks to peter ("to hell with it, it looks ok to me.") for his review.
Thanks to msmith for keeping me from putting locks everywhere :)
Reviewed by: peter
1999-09-19 17:00:25 +00:00
|
|
|
KASSERT(uio->uio_procp == p, ("uio_procp %p is not p %p",
|
|
|
|
|
uio->uio_procp, p));
|
|
|
|
|
vp = (struct vnode *)fp->f_data;
|
1999-04-21 05:56:45 +00:00
|
|
|
ioflag = 0;
|
|
|
|
|
if (fp->f_flag & FNONBLOCK)
|
|
|
|
|
ioflag |= IO_NDELAY;
|
1997-02-10 02:22:35 +00:00
|
|
|
VOP_LEASE(vp, p, cred, LEASE_READ);
|
1997-10-06 02:38:30 +00:00
|
|
|
vn_lock(vp, LK_SHARED | LK_NOPAUSE | LK_RETRY, p);
|
1999-04-21 05:56:45 +00:00
|
|
|
if ((flags & FOF_OFFSET) == 0)
|
|
|
|
|
uio->uio_offset = fp->f_offset;
|
1999-03-26 20:25:21 +00:00
|
|
|
|
2000-04-02 00:55:28 +00:00
|
|
|
ioflag |= sequential_heuristic(uio, fp);
|
|
|
|
|
|
1999-04-21 05:56:45 +00:00
|
|
|
error = VOP_READ(vp, uio, ioflag, cred);
|
|
|
|
|
if ((flags & FOF_OFFSET) == 0)
|
|
|
|
|
fp->f_offset = uio->uio_offset;
|
2000-04-02 00:55:28 +00:00
|
|
|
fp->f_nextoff = uio->uio_offset;
|
1997-02-10 02:22:35 +00:00
|
|
|
VOP_UNLOCK(vp, 0, p);
|
1994-05-24 10:09:53 +00:00
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* File table vnode write routine.
|
|
|
|
|
*/
|
1995-12-17 21:23:44 +00:00
|
|
|
static int
|
This is what was "fdfix2.patch," a fix for fd sharing. It's pretty
far-reaching in fd-land, so you'll want to consult the code for
changes. The biggest change is that now, you don't use
fp->f_ops->fo_foo(fp, bar)
but instead
fo_foo(fp, bar),
which increments and decrements the fp refcount upon entry and exit.
Two new calls, fhold() and fdrop(), are provided. Each does what it
seems like it should, and if fdrop() brings the refcount to zero, the
fd is freed as well.
Thanks to peter ("to hell with it, it looks ok to me.") for his review.
Thanks to msmith for keeping me from putting locks everywhere :)
Reviewed by: peter
1999-09-19 17:00:25 +00:00
|
|
|
vn_write(fp, uio, cred, flags, p)
|
1994-05-24 10:09:53 +00:00
|
|
|
struct file *fp;
|
|
|
|
|
struct uio *uio;
|
|
|
|
|
struct ucred *cred;
|
This is what was "fdfix2.patch," a fix for fd sharing. It's pretty
far-reaching in fd-land, so you'll want to consult the code for
changes. The biggest change is that now, you don't use
fp->f_ops->fo_foo(fp, bar)
but instead
fo_foo(fp, bar),
which increments and decrements the fp refcount upon entry and exit.
Two new calls, fhold() and fdrop(), are provided. Each does what it
seems like it should, and if fdrop() brings the refcount to zero, the
fd is freed as well.
Thanks to peter ("to hell with it, it looks ok to me.") for his review.
Thanks to msmith for keeping me from putting locks everywhere :)
Reviewed by: peter
1999-09-19 17:00:25 +00:00
|
|
|
struct proc *p;
|
1999-04-04 21:41:28 +00:00
|
|
|
int flags;
|
1994-05-24 10:09:53 +00:00
|
|
|
{
|
1999-07-08 06:06:00 +00:00
|
|
|
struct vnode *vp;
|
2000-07-11 22:07:57 +00:00
|
|
|
struct mount *mp;
|
1999-04-21 05:56:45 +00:00
|
|
|
int error, ioflag;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
This is what was "fdfix2.patch," a fix for fd sharing. It's pretty
far-reaching in fd-land, so you'll want to consult the code for
changes. The biggest change is that now, you don't use
fp->f_ops->fo_foo(fp, bar)
but instead
fo_foo(fp, bar),
which increments and decrements the fp refcount upon entry and exit.
Two new calls, fhold() and fdrop(), are provided. Each does what it
seems like it should, and if fdrop() brings the refcount to zero, the
fd is freed as well.
Thanks to peter ("to hell with it, it looks ok to me.") for his review.
Thanks to msmith for keeping me from putting locks everywhere :)
Reviewed by: peter
1999-09-19 17:00:25 +00:00
|
|
|
KASSERT(uio->uio_procp == p, ("uio_procp %p is not p %p",
|
|
|
|
|
uio->uio_procp, p));
|
1999-07-08 06:06:00 +00:00
|
|
|
vp = (struct vnode *)fp->f_data;
|
|
|
|
|
if (vp->v_type == VREG)
|
|
|
|
|
bwillwrite();
|
|
|
|
|
vp = (struct vnode *)fp->f_data; /* XXX needed? */
|
1999-04-21 05:56:45 +00:00
|
|
|
ioflag = IO_UNIT;
|
|
|
|
|
if (vp->v_type == VREG && (fp->f_flag & O_APPEND))
|
1994-05-24 10:09:53 +00:00
|
|
|
ioflag |= IO_APPEND;
|
|
|
|
|
if (fp->f_flag & FNONBLOCK)
|
|
|
|
|
ioflag |= IO_NDELAY;
|
1997-02-10 02:22:35 +00:00
|
|
|
if ((fp->f_flag & O_FSYNC) ||
|
|
|
|
|
(vp->v_mount && (vp->v_mount->mnt_flag & MNT_SYNCHRONOUS)))
|
|
|
|
|
ioflag |= IO_SYNC;
|
2000-07-11 22:07:57 +00:00
|
|
|
mp = NULL;
|
2000-11-02 21:14:13 +00:00
|
|
|
if (vp->v_type != VCHR &&
|
2000-07-11 22:07:57 +00:00
|
|
|
(error = vn_start_write(vp, &mp, V_WAIT | PCATCH)) != 0)
|
|
|
|
|
return (error);
|
1997-02-10 02:22:35 +00:00
|
|
|
VOP_LEASE(vp, p, cred, LEASE_WRITE);
|
|
|
|
|
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, p);
|
1999-04-21 05:56:45 +00:00
|
|
|
if ((flags & FOF_OFFSET) == 0)
|
|
|
|
|
uio->uio_offset = fp->f_offset;
|
2000-04-02 00:55:28 +00:00
|
|
|
ioflag |= sequential_heuristic(uio, fp);
|
1994-05-24 10:09:53 +00:00
|
|
|
error = VOP_WRITE(vp, uio, ioflag, cred);
|
1999-04-21 05:56:45 +00:00
|
|
|
if ((flags & FOF_OFFSET) == 0)
|
1994-05-24 10:09:53 +00:00
|
|
|
fp->f_offset = uio->uio_offset;
|
2000-04-02 00:55:28 +00:00
|
|
|
fp->f_nextoff = uio->uio_offset;
|
1997-02-10 02:22:35 +00:00
|
|
|
VOP_UNLOCK(vp, 0, p);
|
2000-07-11 22:07:57 +00:00
|
|
|
vn_finished_write(mp);
|
1994-05-24 10:09:53 +00:00
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* File table vnode stat routine.
|
|
|
|
|
*/
|
1999-11-08 03:32:15 +00:00
|
|
|
static int
|
|
|
|
|
vn_statfile(fp, sb, p)
|
|
|
|
|
struct file *fp;
|
|
|
|
|
struct stat *sb;
|
|
|
|
|
struct proc *p;
|
|
|
|
|
{
|
|
|
|
|
struct vnode *vp = (struct vnode *)fp->f_data;
|
|
|
|
|
|
|
|
|
|
return vn_stat(vp, sb, p);
|
|
|
|
|
}
|
|
|
|
|
|
1994-05-25 09:21:21 +00:00
|
|
|
int
|
1994-05-24 10:09:53 +00:00
|
|
|
vn_stat(vp, sb, p)
|
|
|
|
|
struct vnode *vp;
|
|
|
|
|
register struct stat *sb;
|
|
|
|
|
struct proc *p;
|
|
|
|
|
{
|
|
|
|
|
struct vattr vattr;
|
|
|
|
|
register struct vattr *vap;
|
|
|
|
|
int error;
|
|
|
|
|
u_short mode;
|
|
|
|
|
|
|
|
|
|
vap = &vattr;
|
|
|
|
|
error = VOP_GETATTR(vp, vap, p->p_ucred, p);
|
|
|
|
|
if (error)
|
|
|
|
|
return (error);
|
1999-11-18 08:14:20 +00:00
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Zero the spare stat fields
|
|
|
|
|
*/
|
|
|
|
|
sb->st_lspare = 0;
|
|
|
|
|
sb->st_qspare[0] = 0;
|
|
|
|
|
sb->st_qspare[1] = 0;
|
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
|
|
|
|
* Copy from vattr table
|
|
|
|
|
*/
|
1999-07-02 16:29:47 +00:00
|
|
|
if (vap->va_fsid != VNOVAL)
|
|
|
|
|
sb->st_dev = vap->va_fsid;
|
|
|
|
|
else
|
|
|
|
|
sb->st_dev = vp->v_mount->mnt_stat.f_fsid.val[0];
|
1994-05-24 10:09:53 +00:00
|
|
|
sb->st_ino = vap->va_fileid;
|
|
|
|
|
mode = vap->va_mode;
|
1998-06-27 06:43:09 +00:00
|
|
|
switch (vap->va_type) {
|
1994-05-24 10:09:53 +00:00
|
|
|
case VREG:
|
|
|
|
|
mode |= S_IFREG;
|
|
|
|
|
break;
|
|
|
|
|
case VDIR:
|
|
|
|
|
mode |= S_IFDIR;
|
|
|
|
|
break;
|
|
|
|
|
case VBLK:
|
|
|
|
|
mode |= S_IFBLK;
|
|
|
|
|
break;
|
|
|
|
|
case VCHR:
|
|
|
|
|
mode |= S_IFCHR;
|
|
|
|
|
break;
|
|
|
|
|
case VLNK:
|
|
|
|
|
mode |= S_IFLNK;
|
1998-04-10 00:09:04 +00:00
|
|
|
/* This is a cosmetic change, symlinks do not have a mode. */
|
1998-04-08 18:31:59 +00:00
|
|
|
if (vp->v_mount->mnt_flag & MNT_NOSYMFOLLOW)
|
|
|
|
|
sb->st_mode &= ~ACCESSPERMS; /* 0000 */
|
|
|
|
|
else
|
|
|
|
|
sb->st_mode |= ACCESSPERMS; /* 0777 */
|
1994-05-24 10:09:53 +00:00
|
|
|
break;
|
|
|
|
|
case VSOCK:
|
|
|
|
|
mode |= S_IFSOCK;
|
|
|
|
|
break;
|
|
|
|
|
case VFIFO:
|
|
|
|
|
mode |= S_IFIFO;
|
|
|
|
|
break;
|
|
|
|
|
default:
|
|
|
|
|
return (EBADF);
|
|
|
|
|
};
|
|
|
|
|
sb->st_mode = mode;
|
|
|
|
|
sb->st_nlink = vap->va_nlink;
|
|
|
|
|
sb->st_uid = vap->va_uid;
|
|
|
|
|
sb->st_gid = vap->va_gid;
|
|
|
|
|
sb->st_rdev = vap->va_rdev;
|
|
|
|
|
sb->st_size = vap->va_size;
|
|
|
|
|
sb->st_atimespec = vap->va_atime;
|
1997-02-10 02:22:35 +00:00
|
|
|
sb->st_mtimespec = vap->va_mtime;
|
1994-05-24 10:09:53 +00:00
|
|
|
sb->st_ctimespec = vap->va_ctime;
|
1999-09-09 19:08:44 +00:00
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* According to www.opengroup.org, the meaning of st_blksize is
|
|
|
|
|
* "a filesystem-specific preferred I/O block size for this
|
|
|
|
|
* object. In some filesystem types, this may vary from file
|
|
|
|
|
* to file"
|
|
|
|
|
* Default to zero to catch bogus uses of this field.
|
1999-08-13 10:56:07 +00:00
|
|
|
*/
|
1999-09-09 19:08:44 +00:00
|
|
|
|
|
|
|
|
if (vap->va_type == VREG) {
|
1999-09-03 05:16:59 +00:00
|
|
|
sb->st_blksize = vap->va_blocksize;
|
2000-01-10 12:04:27 +00:00
|
|
|
} else if (vn_isdisk(vp, NULL)) {
|
1999-09-09 19:08:44 +00:00
|
|
|
sb->st_blksize = vp->v_rdev->si_bsize_best;
|
|
|
|
|
if (sb->st_blksize < vp->v_rdev->si_bsize_phys)
|
|
|
|
|
sb->st_blksize = vp->v_rdev->si_bsize_phys;
|
|
|
|
|
if (sb->st_blksize < BLKDEV_IOSIZE)
|
|
|
|
|
sb->st_blksize = BLKDEV_IOSIZE;
|
|
|
|
|
} else {
|
|
|
|
|
sb->st_blksize = 0;
|
1999-09-03 05:16:59 +00:00
|
|
|
}
|
1999-09-09 19:08:44 +00:00
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
sb->st_flags = vap->va_flags;
|
This Implements the mumbled about "Jail" feature.
This is a seriously beefed up chroot kind of thing. The process
is jailed along the same lines as a chroot does it, but with
additional tough restrictions imposed on what the superuser can do.
For all I know, it is safe to hand over the root bit inside a
prison to the customer living in that prison, this is what
it was developed for in fact: "real virtual servers".
Each prison has an ip number associated with it, which all IP
communications will be coerced to use and each prison has its own
hostname.
Needless to say, you need more RAM this way, but the advantage is
that each customer can run their own particular version of apache
and not stomp on the toes of their neighbors.
It generally does what one would expect, but setting up a jail
still takes a little knowledge.
A few notes:
I have no scripts for setting up a jail, don't ask me for them.
The IP number should be an alias on one of the interfaces.
mount a /proc in each jail, it will make ps more useable.
/proc/<pid>/status tells the hostname of the prison for
jailed processes.
Quotas are only sensible if you have a mountpoint per prison.
There are no privisions for stopping resource-hogging.
Some "#ifdef INET" and similar may be missing (send patches!)
If somebody wants to take it from here and develop it into
more of a "virtual machine" they should be most welcome!
Tools, comments, patches & documentation most welcome.
Have fun...
Sponsored by: http://www.rndassociates.com/
Run for almost a year by: http://www.servetheweb.com/
1999-04-28 11:38:52 +00:00
|
|
|
if (suser_xxx(p->p_ucred, 0, 0))
|
1997-03-08 15:14:30 +00:00
|
|
|
sb->st_gen = 0;
|
|
|
|
|
else
|
|
|
|
|
sb->st_gen = vap->va_gen;
|
1997-03-07 07:42:41 +00:00
|
|
|
|
1995-10-06 09:43:32 +00:00
|
|
|
#if (S_BLKSIZE == 512)
|
|
|
|
|
/* Optimize this case */
|
|
|
|
|
sb->st_blocks = vap->va_bytes >> 9;
|
|
|
|
|
#else
|
1994-05-24 10:09:53 +00:00
|
|
|
sb->st_blocks = vap->va_bytes / S_BLKSIZE;
|
1995-10-06 09:43:32 +00:00
|
|
|
#endif
|
1994-05-24 10:09:53 +00:00
|
|
|
return (0);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* File table vnode ioctl routine.
|
|
|
|
|
*/
|
1995-12-17 21:23:44 +00:00
|
|
|
static int
|
1994-05-24 10:09:53 +00:00
|
|
|
vn_ioctl(fp, com, data, p)
|
|
|
|
|
struct file *fp;
|
1998-06-07 17:13:14 +00:00
|
|
|
u_long com;
|
1994-05-24 10:09:53 +00:00
|
|
|
caddr_t data;
|
|
|
|
|
struct proc *p;
|
|
|
|
|
{
|
|
|
|
|
register struct vnode *vp = ((struct vnode *)fp->f_data);
|
|
|
|
|
struct vattr vattr;
|
|
|
|
|
int error;
|
|
|
|
|
|
|
|
|
|
switch (vp->v_type) {
|
|
|
|
|
|
|
|
|
|
case VREG:
|
|
|
|
|
case VDIR:
|
|
|
|
|
if (com == FIONREAD) {
|
1994-10-02 17:35:40 +00:00
|
|
|
error = VOP_GETATTR(vp, &vattr, p->p_ucred, p);
|
|
|
|
|
if (error)
|
1994-05-24 10:09:53 +00:00
|
|
|
return (error);
|
|
|
|
|
*(int *)data = vattr.va_size - fp->f_offset;
|
|
|
|
|
return (0);
|
|
|
|
|
}
|
|
|
|
|
if (com == FIONBIO || com == FIOASYNC) /* XXX */
|
|
|
|
|
return (0); /* XXX */
|
|
|
|
|
/* fall into ... */
|
|
|
|
|
|
|
|
|
|
default:
|
1997-12-06 04:11:14 +00:00
|
|
|
#if 0
|
1994-05-24 10:09:53 +00:00
|
|
|
return (ENOTTY);
|
1997-12-06 04:11:14 +00:00
|
|
|
#endif
|
1994-05-24 10:09:53 +00:00
|
|
|
case VFIFO:
|
|
|
|
|
case VCHR:
|
|
|
|
|
case VBLK:
|
1999-08-27 16:35:37 +00:00
|
|
|
if (com == FIODTYPE) {
|
|
|
|
|
if (vp->v_type != VCHR && vp->v_type != VBLK)
|
|
|
|
|
return (ENOTTY);
|
|
|
|
|
*(int *)data = devsw(vp->v_rdev)->d_flags & D_TYPEMASK;
|
|
|
|
|
return (0);
|
|
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
error = VOP_IOCTL(vp, com, data, fp->f_flag, p->p_ucred, p);
|
|
|
|
|
if (error == 0 && com == TIOCSCTTY) {
|
1994-08-18 03:53:38 +00:00
|
|
|
|
|
|
|
|
/* Do nothing if reassigning same control tty */
|
|
|
|
|
if (p->p_session->s_ttyvp == vp)
|
|
|
|
|
return (0);
|
|
|
|
|
|
|
|
|
|
/* Get rid of reference to old control tty */
|
|
|
|
|
if (p->p_session->s_ttyvp)
|
|
|
|
|
vrele(p->p_session->s_ttyvp);
|
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
p->p_session->s_ttyvp = vp;
|
|
|
|
|
VREF(vp);
|
|
|
|
|
}
|
|
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
1997-09-14 02:51:16 +00:00
|
|
|
* File table vnode poll routine.
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
1995-12-17 21:23:44 +00:00
|
|
|
static int
|
1997-09-14 02:51:16 +00:00
|
|
|
vn_poll(fp, events, cred, p)
|
1994-05-24 10:09:53 +00:00
|
|
|
struct file *fp;
|
1997-09-14 02:51:16 +00:00
|
|
|
int events;
|
|
|
|
|
struct ucred *cred;
|
1994-05-24 10:09:53 +00:00
|
|
|
struct proc *p;
|
|
|
|
|
{
|
|
|
|
|
|
1997-09-14 02:51:16 +00:00
|
|
|
return (VOP_POLL(((struct vnode *)fp->f_data), events, cred, p));
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
|
|
1997-02-10 02:22:35 +00:00
|
|
|
/*
|
|
|
|
|
* Check that the vnode is still valid, and if so
|
|
|
|
|
* acquire requested lock.
|
|
|
|
|
*/
|
|
|
|
|
int
|
1999-01-20 14:49:12 +00:00
|
|
|
#ifndef DEBUG_LOCKS
|
1997-02-10 02:22:35 +00:00
|
|
|
vn_lock(vp, flags, p)
|
1999-01-20 14:49:12 +00:00
|
|
|
#else
|
|
|
|
|
debug_vn_lock(vp, flags, p, filename, line)
|
|
|
|
|
#endif
|
1997-02-10 02:22:35 +00:00
|
|
|
struct vnode *vp;
|
|
|
|
|
int flags;
|
|
|
|
|
struct proc *p;
|
1999-01-20 14:49:12 +00:00
|
|
|
#ifdef DEBUG_LOCKS
|
|
|
|
|
const char *filename;
|
|
|
|
|
int line;
|
|
|
|
|
#endif
|
1997-02-10 02:22:35 +00:00
|
|
|
{
|
|
|
|
|
int error;
|
2001-03-26 12:45:35 +00:00
|
|
|
|
1997-02-10 02:22:35 +00:00
|
|
|
do {
|
1997-10-27 15:26:23 +00:00
|
|
|
if ((flags & LK_INTERLOCK) == 0)
|
Change and clean the mutex lock interface.
mtx_enter(lock, type) becomes:
mtx_lock(lock) for sleep locks (MTX_DEF-initialized locks)
mtx_lock_spin(lock) for spin locks (MTX_SPIN-initialized)
similarily, for releasing a lock, we now have:
mtx_unlock(lock) for MTX_DEF and mtx_unlock_spin(lock) for MTX_SPIN.
We change the caller interface for the two different types of locks
because the semantics are entirely different for each case, and this
makes it explicitly clear and, at the same time, it rids us of the
extra `type' argument.
The enter->lock and exit->unlock change has been made with the idea
that we're "locking data" and not "entering locked code" in mind.
Further, remove all additional "flags" previously passed to the
lock acquire/release routines with the exception of two:
MTX_QUIET and MTX_NOSWITCH
The functionality of these flags is preserved and they can be passed
to the lock/unlock routines by calling the corresponding wrappers:
mtx_{lock, unlock}_flags(lock, flag(s)) and
mtx_{lock, unlock}_spin_flags(lock, flag(s)) for MTX_DEF and MTX_SPIN
locks, respectively.
Re-inline some lock acq/rel code; in the sleep lock case, we only
inline the _obtain_lock()s in order to ensure that the inlined code
fits into a cache line. In the spin lock case, we inline recursion and
actually only perform a function call if we need to spin. This change
has been made with the idea that we generally tend to avoid spin locks
and that also the spin locks that we do have and are heavily used
(i.e. sched_lock) do recurse, and therefore in an effort to reduce
function call overhead for some architectures (such as alpha), we
inline recursion for this case.
Create a new malloc type for the witness code and retire from using
the M_DEV type. The new type is called M_WITNESS and is only declared
if WITNESS is enabled.
Begin cleaning up some machdep/mutex.h code - specifically updated the
"optimized" inlined code in alpha/mutex.h and wrote MTX_LOCK_SPIN
and MTX_UNLOCK_SPIN asm macros for the i386/mutex.h as we presently
need those.
Finally, caught up to the interface changes in all sys code.
Contributors: jake, jhb, jasone (in no particular order)
2001-02-09 06:11:45 +00:00
|
|
|
mtx_lock(&vp->v_interlock);
|
Implement a low-memory deadlock solution.
Removed most of the hacks that were trying to deal with low-memory
situations prior to now.
The new code is based on the concept that I/O must be able to function in
a low memory situation. All major modules related to I/O (except
networking) have been adjusted to allow allocation out of the system
reserve memory pool. These modules now detect a low memory situation but
rather then block they instead continue to operate, then return resources
to the memory pool instead of cache them or leave them wired.
Code has been added to stall in a low-memory situation prior to a vnode
being locked.
Thus situations where a process blocks in a low-memory condition while
holding a locked vnode have been reduced to near nothing. Not only will
I/O continue to operate, but many prior deadlock conditions simply no
longer exist.
Implement a number of VFS/BIO fixes
(found by Ian): in biodone(), bogus-page replacement code, the loop
was not properly incrementing loop variables prior to a continue
statement. We do not believe this code can be hit anyway but we
aren't taking any chances. We'll turn the whole section into a
panic (as it already is in brelse()) after the release is rolled.
In biodone(), the foff calculation was incorrectly
clamped to the iosize, causing the wrong foff to be calculated
for pages in the case of an I/O error or biodone() called without
initiating I/O. The problem always caused a panic before. Now it
doesn't. The problem is mainly an issue with NFS.
Fixed casts for ~PAGE_MASK. This code worked properly before only
because the calculations use signed arithmatic. Better to properly
extend PAGE_MASK first before inverting it for the 64 bit masking
op.
In brelse(), the bogus_page fixup code was improperly throwing
away the original contents of 'm' when it did the j-loop to
fix the bogus pages. The result was that it would potentially
invalidate parts of the *WRONG* page(!), leading to corruption.
There may still be cases where a background bitmap write is
being duplicated, causing potential corruption. We have identified
a potentially serious bug related to this but the fix is still TBD.
So instead this patch contains a KASSERT to detect the problem
and panic the machine rather then continue to corrupt the filesystem.
The problem does not occur very often.. it is very hard to
reproduce, and it may or may not be the cause of the corruption
people have reported.
Review by: (VFS/BIO: mckusick, Ian Dowse <iedowse@maths.tcd.ie>)
Testing by: (VM/Deadlock) Paul Saab <ps@yahoo-inc.com>
2000-11-18 23:06:26 +00:00
|
|
|
if ((vp->v_flag & VXLOCK) && vp->v_vxproc != curproc) {
|
1997-02-10 02:22:35 +00:00
|
|
|
vp->v_flag |= VXWANT;
|
2001-03-26 12:45:35 +00:00
|
|
|
msleep(vp, &vp->v_interlock, PINOD | PDROP,
|
|
|
|
|
"vn_lock", 0);
|
1997-02-10 02:22:35 +00:00
|
|
|
error = ENOENT;
|
|
|
|
|
} else {
|
Implement a low-memory deadlock solution.
Removed most of the hacks that were trying to deal with low-memory
situations prior to now.
The new code is based on the concept that I/O must be able to function in
a low memory situation. All major modules related to I/O (except
networking) have been adjusted to allow allocation out of the system
reserve memory pool. These modules now detect a low memory situation but
rather then block they instead continue to operate, then return resources
to the memory pool instead of cache them or leave them wired.
Code has been added to stall in a low-memory situation prior to a vnode
being locked.
Thus situations where a process blocks in a low-memory condition while
holding a locked vnode have been reduced to near nothing. Not only will
I/O continue to operate, but many prior deadlock conditions simply no
longer exist.
Implement a number of VFS/BIO fixes
(found by Ian): in biodone(), bogus-page replacement code, the loop
was not properly incrementing loop variables prior to a continue
statement. We do not believe this code can be hit anyway but we
aren't taking any chances. We'll turn the whole section into a
panic (as it already is in brelse()) after the release is rolled.
In biodone(), the foff calculation was incorrectly
clamped to the iosize, causing the wrong foff to be calculated
for pages in the case of an I/O error or biodone() called without
initiating I/O. The problem always caused a panic before. Now it
doesn't. The problem is mainly an issue with NFS.
Fixed casts for ~PAGE_MASK. This code worked properly before only
because the calculations use signed arithmatic. Better to properly
extend PAGE_MASK first before inverting it for the 64 bit masking
op.
In brelse(), the bogus_page fixup code was improperly throwing
away the original contents of 'm' when it did the j-loop to
fix the bogus pages. The result was that it would potentially
invalidate parts of the *WRONG* page(!), leading to corruption.
There may still be cases where a background bitmap write is
being duplicated, causing potential corruption. We have identified
a potentially serious bug related to this but the fix is still TBD.
So instead this patch contains a KASSERT to detect the problem
and panic the machine rather then continue to corrupt the filesystem.
The problem does not occur very often.. it is very hard to
reproduce, and it may or may not be the cause of the corruption
people have reported.
Review by: (VFS/BIO: mckusick, Ian Dowse <iedowse@maths.tcd.ie>)
Testing by: (VM/Deadlock) Paul Saab <ps@yahoo-inc.com>
2000-11-18 23:06:26 +00:00
|
|
|
if (vp->v_vxproc != NULL)
|
|
|
|
|
printf("VXLOCK interlock avoided in vn_lock\n");
|
1999-01-20 14:49:12 +00:00
|
|
|
#ifdef DEBUG_LOCKS
|
|
|
|
|
vp->filename = filename;
|
|
|
|
|
vp->line = line;
|
|
|
|
|
#endif
|
|
|
|
|
error = VOP_LOCK(vp,
|
|
|
|
|
flags | LK_NOPAUSE | LK_INTERLOCK, p);
|
1997-02-10 02:22:35 +00:00
|
|
|
if (error == 0)
|
|
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
flags &= ~LK_INTERLOCK;
|
|
|
|
|
} while (flags & LK_RETRY);
|
|
|
|
|
return (error);
|
|
|
|
|
}
|
1997-10-27 15:26:23 +00:00
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* File table vnode close routine.
|
|
|
|
|
*/
|
|
|
|
|
static int
|
|
|
|
|
vn_closefile(fp, p)
|
|
|
|
|
struct file *fp;
|
|
|
|
|
struct proc *p;
|
|
|
|
|
{
|
|
|
|
|
|
1999-08-04 18:53:50 +00:00
|
|
|
fp->f_ops = &badfileops;
|
1997-10-27 15:26:23 +00:00
|
|
|
return (vn_close(((struct vnode *)fp->f_data), fp->f_flag,
|
|
|
|
|
fp->f_cred, p));
|
|
|
|
|
}
|
2000-04-16 18:53:38 +00:00
|
|
|
|
2000-07-11 22:07:57 +00:00
|
|
|
/*
|
|
|
|
|
* Preparing to start a filesystem write operation. If the operation is
|
|
|
|
|
* permitted, then we bump the count of operations in progress and
|
|
|
|
|
* proceed. If a suspend request is in progress, we wait until the
|
|
|
|
|
* suspension is over, and then proceed.
|
|
|
|
|
*/
|
|
|
|
|
int
|
|
|
|
|
vn_start_write(vp, mpp, flags)
|
|
|
|
|
struct vnode *vp;
|
|
|
|
|
struct mount **mpp;
|
|
|
|
|
int flags;
|
|
|
|
|
{
|
|
|
|
|
struct mount *mp;
|
|
|
|
|
int error;
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* If a vnode is provided, get and return the mount point that
|
|
|
|
|
* to which it will write.
|
|
|
|
|
*/
|
|
|
|
|
if (vp != NULL) {
|
|
|
|
|
if ((error = VOP_GETWRITEMOUNT(vp, mpp)) != 0) {
|
|
|
|
|
*mpp = NULL;
|
|
|
|
|
if (error != EOPNOTSUPP)
|
|
|
|
|
return (error);
|
|
|
|
|
return (0);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
if ((mp = *mpp) == NULL)
|
|
|
|
|
return (0);
|
|
|
|
|
/*
|
|
|
|
|
* Check on status of suspension.
|
|
|
|
|
*/
|
|
|
|
|
while ((mp->mnt_kern_flag & MNTK_SUSPEND) != 0) {
|
|
|
|
|
if (flags & V_NOWAIT)
|
|
|
|
|
return (EWOULDBLOCK);
|
|
|
|
|
error = tsleep(&mp->mnt_flag, (PUSER - 1) | (flags & PCATCH),
|
|
|
|
|
"suspfs", 0);
|
|
|
|
|
if (error)
|
|
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
if (flags & V_XSLEEP)
|
|
|
|
|
return (0);
|
|
|
|
|
mp->mnt_writeopcount++;
|
|
|
|
|
return (0);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Secondary suspension. Used by operations such as vop_inactive
|
|
|
|
|
* routines that are needed by the higher level functions. These
|
|
|
|
|
* are allowed to proceed until all the higher level functions have
|
|
|
|
|
* completed (indicated by mnt_writeopcount dropping to zero). At that
|
|
|
|
|
* time, these operations are halted until the suspension is over.
|
|
|
|
|
*/
|
|
|
|
|
int
|
2000-07-24 05:28:33 +00:00
|
|
|
vn_write_suspend_wait(vp, mp, flags)
|
2000-07-11 22:07:57 +00:00
|
|
|
struct vnode *vp;
|
2000-07-24 05:28:33 +00:00
|
|
|
struct mount *mp;
|
2000-07-11 22:07:57 +00:00
|
|
|
int flags;
|
|
|
|
|
{
|
|
|
|
|
int error;
|
|
|
|
|
|
2000-07-24 05:28:33 +00:00
|
|
|
if (vp != NULL) {
|
|
|
|
|
if ((error = VOP_GETWRITEMOUNT(vp, &mp)) != 0) {
|
|
|
|
|
if (error != EOPNOTSUPP)
|
|
|
|
|
return (error);
|
|
|
|
|
return (0);
|
|
|
|
|
}
|
2000-07-11 22:07:57 +00:00
|
|
|
}
|
|
|
|
|
/*
|
|
|
|
|
* If we are not suspended or have not yet reached suspended
|
|
|
|
|
* mode, then let the operation proceed.
|
|
|
|
|
*/
|
|
|
|
|
if (mp == NULL || (mp->mnt_kern_flag & MNTK_SUSPENDED) == 0)
|
|
|
|
|
return (0);
|
|
|
|
|
if (flags & V_NOWAIT)
|
|
|
|
|
return (EWOULDBLOCK);
|
|
|
|
|
/*
|
|
|
|
|
* Wait for the suspension to finish.
|
|
|
|
|
*/
|
|
|
|
|
return (tsleep(&mp->mnt_flag, (PUSER - 1) | (flags & PCATCH),
|
|
|
|
|
"suspfs", 0));
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Filesystem write operation has completed. If we are suspending and this
|
|
|
|
|
* operation is the last one, notify the suspender that the suspension is
|
|
|
|
|
* now in effect.
|
|
|
|
|
*/
|
|
|
|
|
void
|
|
|
|
|
vn_finished_write(mp)
|
|
|
|
|
struct mount *mp;
|
|
|
|
|
{
|
|
|
|
|
|
|
|
|
|
if (mp == NULL)
|
|
|
|
|
return;
|
|
|
|
|
mp->mnt_writeopcount--;
|
|
|
|
|
if (mp->mnt_writeopcount < 0)
|
|
|
|
|
panic("vn_finished_write: neg cnt");
|
|
|
|
|
if ((mp->mnt_kern_flag & MNTK_SUSPEND) != 0 &&
|
|
|
|
|
mp->mnt_writeopcount <= 0)
|
|
|
|
|
wakeup(&mp->mnt_writeopcount);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Request a filesystem to suspend write operations.
|
|
|
|
|
*/
|
|
|
|
|
void
|
|
|
|
|
vfs_write_suspend(mp)
|
|
|
|
|
struct mount *mp;
|
|
|
|
|
{
|
|
|
|
|
struct proc *p = curproc;
|
|
|
|
|
|
|
|
|
|
if (mp->mnt_kern_flag & MNTK_SUSPEND)
|
|
|
|
|
return;
|
|
|
|
|
mp->mnt_kern_flag |= MNTK_SUSPEND;
|
|
|
|
|
if (mp->mnt_writeopcount > 0)
|
|
|
|
|
(void) tsleep(&mp->mnt_writeopcount, PUSER - 1, "suspwt", 0);
|
|
|
|
|
VFS_SYNC(mp, MNT_WAIT, p->p_ucred, p);
|
|
|
|
|
mp->mnt_kern_flag |= MNTK_SUSPENDED;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Request a filesystem to resume write operations.
|
|
|
|
|
*/
|
|
|
|
|
void
|
|
|
|
|
vfs_write_resume(mp)
|
|
|
|
|
struct mount *mp;
|
|
|
|
|
{
|
|
|
|
|
|
|
|
|
|
if ((mp->mnt_kern_flag & MNTK_SUSPEND) == 0)
|
|
|
|
|
return;
|
|
|
|
|
mp->mnt_kern_flag &= ~(MNTK_SUSPEND | MNTK_SUSPENDED);
|
|
|
|
|
wakeup(&mp->mnt_writeopcount);
|
|
|
|
|
wakeup(&mp->mnt_flag);
|
|
|
|
|
}
|
|
|
|
|
|
2000-04-16 18:53:38 +00:00
|
|
|
static int
|
2001-02-15 16:34:11 +00:00
|
|
|
vn_kqfilter(struct file *fp, struct knote *kn)
|
2000-04-16 18:53:38 +00:00
|
|
|
{
|
|
|
|
|
|
2001-02-15 16:34:11 +00:00
|
|
|
return (VOP_KQFILTER(((struct vnode *)fp->f_data), kn));
|
2000-04-16 18:53:38 +00:00
|
|
|
}
|
2000-08-08 17:15:32 +00:00
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Simplified in-kernel wrapper calls for extended attribute access.
|
|
|
|
|
* Both calls pass in a NULL credential, authorizing as "kernel" access.
|
|
|
|
|
* Set IO_NODELOCKED in ioflg if the vnode is already locked.
|
|
|
|
|
*/
|
|
|
|
|
int
|
2001-03-19 05:44:15 +00:00
|
|
|
vn_extattr_get(struct vnode *vp, int ioflg, int attrnamespace,
|
o Change the API and ABI of the Extended Attribute kernel interfaces to
introduce a new argument, "namespace", rather than relying on a first-
character namespace indicator. This is in line with more recent
thinking on EA interfaces on various mailing lists, including the
posix1e, Linux acl-devel, and trustedbsd-discuss forums. Two namespaces
are defined by default, EXTATTR_NAMESPACE_SYSTEM and
EXTATTR_NAMESPACE_USER, where the primary distinction lies in the
access control model: user EAs are accessible based on the normal
MAC and DAC file/directory protections, and system attributes are
limited to kernel-originated or appropriately privileged userland
requests.
o These API changes occur at several levels: the namespace argument is
introduced in the extattr_{get,set}_file() system call interfaces,
at the vnode operation level in the vop_{get,set}extattr() interfaces,
and in the UFS extended attribute implementation. Changes are also
introduced in the VFS extattrctl() interface (system call, VFS,
and UFS implementation), where the arguments are modified to include
a namespace field, as well as modified to advoid direct access to
userspace variables from below the VFS layer (in the style of recent
changes to mount by adrian@FreeBSD.org). This required some cleanup
and bug fixing regarding VFS locks and the VFS interface, as a vnode
pointer may now be optionally submitted to the VFS_EXTATTRCTL()
call. Updated documentation for the VFS interface will be committed
shortly.
o In the near future, the auto-starting feature will be updated to
search two sub-directories to the ".attribute" directory in appropriate
file systems: "user" and "system" to locate attributes intended for
those namespaces, as the single filename is no longer sufficient
to indicate what namespace the attribute is intended for. Until this
is committed, all attributes auto-started by UFS will be placed in
the EXTATTR_NAMESPACE_SYSTEM namespace.
o The default POSIX.1e attribute names for ACLs and Capabilities have
been updated to no longer include the '$' in their filename. As such,
if you're using these features, you'll need to rename the attribute
backing files to the same names without '$' symbols in front.
o Note that these changes will require changes in userland, which will
be committed shortly. These include modifications to the extended
attribute utilities, as well as to libutil for new namespace
string conversion routines. Once the matching userland changes are
committed, a buildworld is recommended to update all the necessary
include files and verify that the kernel and userland environments
are in sync. Note: If you do not use extended attributes (most people
won't), upgrading is not imperative although since the system call
API has changed, the new userland extended attribute code will no longer
compile with old include files.
o Couple of minor cleanups while I'm there: make more code compilation
conditional on FFS_EXTATTR, which should recover a bit of space on
kernels running without EA's, as well as update copyright dates.
Obtained from: TrustedBSD Project
2001-03-15 02:54:29 +00:00
|
|
|
const char *attrname, int *buflen, char *buf, struct proc *p)
|
2000-08-08 17:15:32 +00:00
|
|
|
{
|
|
|
|
|
struct uio auio;
|
|
|
|
|
struct iovec iov;
|
|
|
|
|
int error;
|
|
|
|
|
|
|
|
|
|
iov.iov_len = *buflen;
|
|
|
|
|
iov.iov_base = buf;
|
|
|
|
|
|
|
|
|
|
auio.uio_iov = &iov;
|
|
|
|
|
auio.uio_iovcnt = 1;
|
|
|
|
|
auio.uio_rw = UIO_READ;
|
|
|
|
|
auio.uio_segflg = UIO_SYSSPACE;
|
|
|
|
|
auio.uio_procp = p;
|
|
|
|
|
auio.uio_offset = 0;
|
|
|
|
|
auio.uio_resid = *buflen;
|
|
|
|
|
|
|
|
|
|
if ((ioflg & IO_NODELOCKED) == 0)
|
|
|
|
|
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, p);
|
|
|
|
|
|
|
|
|
|
/* authorize attribute retrieval as kernel */
|
2001-03-19 05:44:15 +00:00
|
|
|
error = VOP_GETEXTATTR(vp, attrnamespace, attrname, &auio, NULL, p);
|
2000-08-08 17:15:32 +00:00
|
|
|
|
|
|
|
|
if ((ioflg & IO_NODELOCKED) == 0)
|
|
|
|
|
VOP_UNLOCK(vp, 0, p);
|
|
|
|
|
|
|
|
|
|
if (error == 0) {
|
|
|
|
|
*buflen = *buflen - auio.uio_resid;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* XXX failure mode if partially written?
|
|
|
|
|
*/
|
|
|
|
|
int
|
2001-03-19 05:44:15 +00:00
|
|
|
vn_extattr_set(struct vnode *vp, int ioflg, int attrnamespace,
|
o Change the API and ABI of the Extended Attribute kernel interfaces to
introduce a new argument, "namespace", rather than relying on a first-
character namespace indicator. This is in line with more recent
thinking on EA interfaces on various mailing lists, including the
posix1e, Linux acl-devel, and trustedbsd-discuss forums. Two namespaces
are defined by default, EXTATTR_NAMESPACE_SYSTEM and
EXTATTR_NAMESPACE_USER, where the primary distinction lies in the
access control model: user EAs are accessible based on the normal
MAC and DAC file/directory protections, and system attributes are
limited to kernel-originated or appropriately privileged userland
requests.
o These API changes occur at several levels: the namespace argument is
introduced in the extattr_{get,set}_file() system call interfaces,
at the vnode operation level in the vop_{get,set}extattr() interfaces,
and in the UFS extended attribute implementation. Changes are also
introduced in the VFS extattrctl() interface (system call, VFS,
and UFS implementation), where the arguments are modified to include
a namespace field, as well as modified to advoid direct access to
userspace variables from below the VFS layer (in the style of recent
changes to mount by adrian@FreeBSD.org). This required some cleanup
and bug fixing regarding VFS locks and the VFS interface, as a vnode
pointer may now be optionally submitted to the VFS_EXTATTRCTL()
call. Updated documentation for the VFS interface will be committed
shortly.
o In the near future, the auto-starting feature will be updated to
search two sub-directories to the ".attribute" directory in appropriate
file systems: "user" and "system" to locate attributes intended for
those namespaces, as the single filename is no longer sufficient
to indicate what namespace the attribute is intended for. Until this
is committed, all attributes auto-started by UFS will be placed in
the EXTATTR_NAMESPACE_SYSTEM namespace.
o The default POSIX.1e attribute names for ACLs and Capabilities have
been updated to no longer include the '$' in their filename. As such,
if you're using these features, you'll need to rename the attribute
backing files to the same names without '$' symbols in front.
o Note that these changes will require changes in userland, which will
be committed shortly. These include modifications to the extended
attribute utilities, as well as to libutil for new namespace
string conversion routines. Once the matching userland changes are
committed, a buildworld is recommended to update all the necessary
include files and verify that the kernel and userland environments
are in sync. Note: If you do not use extended attributes (most people
won't), upgrading is not imperative although since the system call
API has changed, the new userland extended attribute code will no longer
compile with old include files.
o Couple of minor cleanups while I'm there: make more code compilation
conditional on FFS_EXTATTR, which should recover a bit of space on
kernels running without EA's, as well as update copyright dates.
Obtained from: TrustedBSD Project
2001-03-15 02:54:29 +00:00
|
|
|
const char *attrname, int buflen, char *buf, struct proc *p)
|
2000-08-08 17:15:32 +00:00
|
|
|
{
|
|
|
|
|
struct uio auio;
|
|
|
|
|
struct iovec iov;
|
2000-09-05 03:15:02 +00:00
|
|
|
struct mount *mp;
|
2000-08-08 17:15:32 +00:00
|
|
|
int error;
|
|
|
|
|
|
|
|
|
|
iov.iov_len = buflen;
|
|
|
|
|
iov.iov_base = buf;
|
|
|
|
|
|
|
|
|
|
auio.uio_iov = &iov;
|
|
|
|
|
auio.uio_iovcnt = 1;
|
|
|
|
|
auio.uio_rw = UIO_WRITE;
|
|
|
|
|
auio.uio_segflg = UIO_SYSSPACE;
|
|
|
|
|
auio.uio_procp = p;
|
|
|
|
|
auio.uio_offset = 0;
|
|
|
|
|
auio.uio_resid = buflen;
|
|
|
|
|
|
2000-09-05 03:15:02 +00:00
|
|
|
if ((ioflg & IO_NODELOCKED) == 0) {
|
|
|
|
|
if ((error = vn_start_write(vp, &mp, V_WAIT)) != 0)
|
|
|
|
|
return (error);
|
2000-08-08 17:15:32 +00:00
|
|
|
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, p);
|
2000-09-05 03:15:02 +00:00
|
|
|
}
|
2000-08-08 17:15:32 +00:00
|
|
|
|
|
|
|
|
/* authorize attribute setting as kernel */
|
2001-03-19 05:44:15 +00:00
|
|
|
error = VOP_SETEXTATTR(vp, attrnamespace, attrname, &auio, NULL, p);
|
2000-08-08 17:15:32 +00:00
|
|
|
|
2000-09-05 03:15:02 +00:00
|
|
|
if ((ioflg & IO_NODELOCKED) == 0) {
|
|
|
|
|
vn_finished_write(mp);
|
2000-08-08 17:15:32 +00:00
|
|
|
VOP_UNLOCK(vp, 0, p);
|
2000-09-05 03:15:02 +00:00
|
|
|
}
|
2000-08-08 17:15:32 +00:00
|
|
|
|
|
|
|
|
return (error);
|
|
|
|
|
}
|
2000-09-22 22:33:13 +00:00
|
|
|
|
|
|
|
|
int
|
2001-03-19 05:44:15 +00:00
|
|
|
vn_extattr_rm(struct vnode *vp, int ioflg, int attrnamespace,
|
|
|
|
|
const char *attrname, struct proc *p)
|
2000-09-22 22:33:13 +00:00
|
|
|
{
|
|
|
|
|
struct mount *mp;
|
|
|
|
|
int error;
|
|
|
|
|
|
|
|
|
|
if ((ioflg & IO_NODELOCKED) == 0) {
|
|
|
|
|
if ((error = vn_start_write(vp, &mp, V_WAIT)) != 0)
|
|
|
|
|
return (error);
|
|
|
|
|
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, p);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* authorize attribute removal as kernel */
|
2001-03-19 05:44:15 +00:00
|
|
|
error = VOP_SETEXTATTR(vp, attrnamespace, attrname, NULL, NULL, p);
|
2000-09-22 22:33:13 +00:00
|
|
|
|
|
|
|
|
if ((ioflg & IO_NODELOCKED) == 0) {
|
|
|
|
|
vn_finished_write(mp);
|
|
|
|
|
VOP_UNLOCK(vp, 0, p);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return (error);
|
|
|
|
|
}
|