76951d21d1
structure in the struct pointed to by the 3rd argument for IPC_STAT and get rid of the 4th argument. The old way returned a pointer into the kernel array that the calling function would then access afterwards without holding the appropriate locks and doing non-lock-safe things like copyout() with the data anyways. This change removes that unsafeness and resulting race conditions as well as simplifying the interface. - Implement kern_foo wrappers for stat(), lstat(), fstat(), statfs(), fstatfs(), and fhstatfs(). Use these wrappers to cut out a lot of code duplication for freebsd4 and netbsd compatability system calls. - Add a new lookup function kern_alternate_path() that looks up a filename under an alternate prefix and determines which filename should be used. This is basically a more general version of linux_emul_convpath() that can be shared by all the ABIs thus allowing for further reduction of code duplication.
923 lines
24 KiB
C
923 lines
24 KiB
C
/*-
|
|
* 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.
|
|
* 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_lookup.c 8.4 (Berkeley) 2/16/94
|
|
*/
|
|
|
|
#include <sys/cdefs.h>
|
|
__FBSDID("$FreeBSD$");
|
|
|
|
#include "opt_ktrace.h"
|
|
#include "opt_mac.h"
|
|
|
|
#include <sys/param.h>
|
|
#include <sys/systm.h>
|
|
#include <sys/kernel.h>
|
|
#include <sys/lock.h>
|
|
#include <sys/mac.h>
|
|
#include <sys/mutex.h>
|
|
#include <sys/namei.h>
|
|
#include <sys/vnode.h>
|
|
#include <sys/mount.h>
|
|
#include <sys/filedesc.h>
|
|
#include <sys/proc.h>
|
|
#include <sys/syscallsubr.h>
|
|
#ifdef KTRACE
|
|
#include <sys/ktrace.h>
|
|
#endif
|
|
|
|
#include <vm/uma.h>
|
|
|
|
#define NAMEI_DIAGNOSTIC 1
|
|
#undef NAMEI_DIAGNOSTIC
|
|
|
|
/*
|
|
* Allocation zone for namei
|
|
*/
|
|
uma_zone_t namei_zone;
|
|
|
|
static void
|
|
nameiinit(void *dummy __unused)
|
|
{
|
|
namei_zone = uma_zcreate("NAMEI", MAXPATHLEN, NULL, NULL, NULL, NULL,
|
|
UMA_ALIGN_PTR, 0);
|
|
|
|
}
|
|
SYSINIT(vfs, SI_SUB_VFS, SI_ORDER_SECOND, nameiinit, NULL)
|
|
|
|
/*
|
|
* Convert a pathname into a pointer to a locked inode.
|
|
*
|
|
* The FOLLOW flag is set when symbolic links are to be followed
|
|
* when they occur at the end of the name translation process.
|
|
* Symbolic links are always followed for all other pathname
|
|
* components other than the last.
|
|
*
|
|
* The segflg defines whether the name is to be copied from user
|
|
* space or kernel space.
|
|
*
|
|
* Overall outline of namei:
|
|
*
|
|
* copy in name
|
|
* get starting directory
|
|
* while (!done && !error) {
|
|
* call lookup to search path.
|
|
* if symbolic link, massage name in buffer and continue
|
|
* }
|
|
*/
|
|
int
|
|
namei(ndp)
|
|
register struct nameidata *ndp;
|
|
{
|
|
register struct filedesc *fdp; /* pointer to file descriptor state */
|
|
register char *cp; /* pointer into pathname argument */
|
|
register struct vnode *dp; /* the directory we are searching */
|
|
struct iovec aiov; /* uio for reading symbolic links */
|
|
struct uio auio;
|
|
int error, linklen;
|
|
struct componentname *cnp = &ndp->ni_cnd;
|
|
struct thread *td = cnp->cn_thread;
|
|
struct proc *p = td->td_proc;
|
|
struct mount *mp;
|
|
int vfslocked;
|
|
|
|
ndp->ni_cnd.cn_cred = ndp->ni_cnd.cn_thread->td_ucred;
|
|
KASSERT(cnp->cn_cred && p, ("namei: bad cred/proc"));
|
|
KASSERT((cnp->cn_nameiop & (~OPMASK)) == 0,
|
|
("namei: nameiop contaminated with flags"));
|
|
KASSERT((cnp->cn_flags & OPMASK) == 0,
|
|
("namei: flags contaminated with nameiops"));
|
|
fdp = p->p_fd;
|
|
|
|
/*
|
|
* Get a buffer for the name to be translated, and copy the
|
|
* name into the buffer.
|
|
*/
|
|
if ((cnp->cn_flags & HASBUF) == 0)
|
|
cnp->cn_pnbuf = uma_zalloc(namei_zone, M_WAITOK);
|
|
if (ndp->ni_segflg == UIO_SYSSPACE)
|
|
error = copystr(ndp->ni_dirp, cnp->cn_pnbuf,
|
|
MAXPATHLEN, (size_t *)&ndp->ni_pathlen);
|
|
else
|
|
error = copyinstr(ndp->ni_dirp, cnp->cn_pnbuf,
|
|
MAXPATHLEN, (size_t *)&ndp->ni_pathlen);
|
|
|
|
/*
|
|
* Don't allow empty pathnames.
|
|
*/
|
|
if (!error && *cnp->cn_pnbuf == '\0')
|
|
error = ENOENT;
|
|
|
|
if (error) {
|
|
uma_zfree(namei_zone, cnp->cn_pnbuf);
|
|
#ifdef DIAGNOSTIC
|
|
cnp->cn_pnbuf = NULL;
|
|
cnp->cn_nameptr = NULL;
|
|
#endif
|
|
ndp->ni_vp = NULL;
|
|
return (error);
|
|
}
|
|
ndp->ni_loopcnt = 0;
|
|
#ifdef KTRACE
|
|
if (KTRPOINT(td, KTR_NAMEI)) {
|
|
KASSERT(cnp->cn_thread == curthread,
|
|
("namei not using curthread"));
|
|
ktrnamei(cnp->cn_pnbuf);
|
|
}
|
|
#endif
|
|
|
|
/*
|
|
* Get starting point for the translation.
|
|
*/
|
|
FILEDESC_LOCK(fdp);
|
|
ndp->ni_rootdir = fdp->fd_rdir;
|
|
ndp->ni_topdir = fdp->fd_jdir;
|
|
|
|
dp = fdp->fd_cdir;
|
|
vfslocked = VFS_LOCK_GIANT(dp->v_mount);
|
|
VREF(dp);
|
|
FILEDESC_UNLOCK(fdp);
|
|
for (;;) {
|
|
/*
|
|
* Check if root directory should replace current directory.
|
|
* Done at start of translation and after symbolic link.
|
|
*/
|
|
cnp->cn_nameptr = cnp->cn_pnbuf;
|
|
if (*(cnp->cn_nameptr) == '/') {
|
|
vrele(dp);
|
|
VFS_UNLOCK_GIANT(vfslocked);
|
|
while (*(cnp->cn_nameptr) == '/') {
|
|
cnp->cn_nameptr++;
|
|
ndp->ni_pathlen--;
|
|
}
|
|
dp = ndp->ni_rootdir;
|
|
vfslocked = VFS_LOCK_GIANT(dp->v_mount);
|
|
VREF(dp);
|
|
}
|
|
if (vfslocked)
|
|
ndp->ni_cnd.cn_flags |= GIANTHELD;
|
|
ndp->ni_startdir = dp;
|
|
error = lookup(ndp);
|
|
if (error) {
|
|
uma_zfree(namei_zone, cnp->cn_pnbuf);
|
|
#ifdef DIAGNOSTIC
|
|
cnp->cn_pnbuf = NULL;
|
|
cnp->cn_nameptr = NULL;
|
|
#endif
|
|
return (error);
|
|
}
|
|
vfslocked = (ndp->ni_cnd.cn_flags & GIANTHELD) != 0;
|
|
ndp->ni_cnd.cn_flags &= ~GIANTHELD;
|
|
/*
|
|
* Check for symbolic link
|
|
*/
|
|
if ((cnp->cn_flags & ISSYMLINK) == 0) {
|
|
if ((cnp->cn_flags & (SAVENAME | SAVESTART)) == 0) {
|
|
uma_zfree(namei_zone, cnp->cn_pnbuf);
|
|
#ifdef DIAGNOSTIC
|
|
cnp->cn_pnbuf = NULL;
|
|
cnp->cn_nameptr = NULL;
|
|
#endif
|
|
} else
|
|
cnp->cn_flags |= HASBUF;
|
|
|
|
if ((cnp->cn_flags & MPSAFE) == 0) {
|
|
VFS_UNLOCK_GIANT(vfslocked);
|
|
} else if (vfslocked)
|
|
ndp->ni_cnd.cn_flags |= GIANTHELD;
|
|
return (0);
|
|
}
|
|
if ((cnp->cn_flags & LOCKPARENT) && ndp->ni_pathlen == 1)
|
|
VOP_UNLOCK(ndp->ni_dvp, 0, td);
|
|
if (ndp->ni_loopcnt++ >= MAXSYMLINKS) {
|
|
error = ELOOP;
|
|
break;
|
|
}
|
|
#ifdef MAC
|
|
if ((cnp->cn_flags & NOMACCHECK) == 0) {
|
|
error = mac_check_vnode_readlink(td->td_ucred,
|
|
ndp->ni_vp);
|
|
if (error)
|
|
break;
|
|
}
|
|
#endif
|
|
if (ndp->ni_pathlen > 1)
|
|
cp = uma_zalloc(namei_zone, M_WAITOK);
|
|
else
|
|
cp = cnp->cn_pnbuf;
|
|
aiov.iov_base = cp;
|
|
aiov.iov_len = MAXPATHLEN;
|
|
auio.uio_iov = &aiov;
|
|
auio.uio_iovcnt = 1;
|
|
auio.uio_offset = 0;
|
|
auio.uio_rw = UIO_READ;
|
|
auio.uio_segflg = UIO_SYSSPACE;
|
|
auio.uio_td = (struct thread *)0;
|
|
auio.uio_resid = MAXPATHLEN;
|
|
error = VOP_READLINK(ndp->ni_vp, &auio, cnp->cn_cred);
|
|
if (error) {
|
|
if (ndp->ni_pathlen > 1)
|
|
uma_zfree(namei_zone, cp);
|
|
break;
|
|
}
|
|
linklen = MAXPATHLEN - auio.uio_resid;
|
|
if (linklen == 0) {
|
|
if (ndp->ni_pathlen > 1)
|
|
uma_zfree(namei_zone, cp);
|
|
error = ENOENT;
|
|
break;
|
|
}
|
|
if (linklen + ndp->ni_pathlen >= MAXPATHLEN) {
|
|
if (ndp->ni_pathlen > 1)
|
|
uma_zfree(namei_zone, cp);
|
|
error = ENAMETOOLONG;
|
|
break;
|
|
}
|
|
if (ndp->ni_pathlen > 1) {
|
|
bcopy(ndp->ni_next, cp + linklen, ndp->ni_pathlen);
|
|
uma_zfree(namei_zone, cnp->cn_pnbuf);
|
|
cnp->cn_pnbuf = cp;
|
|
} else
|
|
cnp->cn_pnbuf[linklen] = '\0';
|
|
ndp->ni_pathlen += linklen;
|
|
vput(ndp->ni_vp);
|
|
dp = ndp->ni_dvp;
|
|
}
|
|
uma_zfree(namei_zone, cnp->cn_pnbuf);
|
|
#ifdef DIAGNOSTIC
|
|
cnp->cn_pnbuf = NULL;
|
|
cnp->cn_nameptr = NULL;
|
|
#endif
|
|
vrele(ndp->ni_dvp);
|
|
mp = ndp->ni_vp->v_mount;
|
|
vput(ndp->ni_vp);
|
|
VFS_UNLOCK_GIANT(vfslocked);
|
|
ndp->ni_vp = NULL;
|
|
return (error);
|
|
}
|
|
|
|
/*
|
|
* Search a pathname.
|
|
* This is a very central and rather complicated routine.
|
|
*
|
|
* The pathname is pointed to by ni_ptr and is of length ni_pathlen.
|
|
* The starting directory is taken from ni_startdir. The pathname is
|
|
* descended until done, or a symbolic link is encountered. The variable
|
|
* ni_more is clear if the path is completed; it is set to one if a
|
|
* symbolic link needing interpretation is encountered.
|
|
*
|
|
* The flag argument is LOOKUP, CREATE, RENAME, or DELETE depending on
|
|
* whether the name is to be looked up, created, renamed, or deleted.
|
|
* When CREATE, RENAME, or DELETE is specified, information usable in
|
|
* creating, renaming, or deleting a directory entry may be calculated.
|
|
* If flag has LOCKPARENT or'ed into it, the parent directory is returned
|
|
* locked. If flag has WANTPARENT or'ed into it, the parent directory is
|
|
* returned unlocked. Otherwise the parent directory is not returned. If
|
|
* the target of the pathname exists and LOCKLEAF is or'ed into the flag
|
|
* the target is returned locked, otherwise it is returned unlocked.
|
|
* When creating or renaming and LOCKPARENT is specified, the target may not
|
|
* be ".". When deleting and LOCKPARENT is specified, the target may be ".".
|
|
*
|
|
* Overall outline of lookup:
|
|
*
|
|
* dirloop:
|
|
* identify next component of name at ndp->ni_ptr
|
|
* handle degenerate case where name is null string
|
|
* if .. and crossing mount points and on mounted filesys, find parent
|
|
* call VOP_LOOKUP routine for next component name
|
|
* directory vnode returned in ni_dvp, unlocked unless LOCKPARENT set
|
|
* component vnode returned in ni_vp (if it exists), locked.
|
|
* if result vnode is mounted on and crossing mount points,
|
|
* find mounted on vnode
|
|
* if more components of name, do next level at dirloop
|
|
* return the answer in ni_vp, locked if LOCKLEAF set
|
|
* if LOCKPARENT set, return locked parent in ni_dvp
|
|
* if WANTPARENT set, return unlocked parent in ni_dvp
|
|
*/
|
|
int
|
|
lookup(ndp)
|
|
register struct nameidata *ndp;
|
|
{
|
|
register char *cp; /* pointer into pathname argument */
|
|
register struct vnode *dp = 0; /* the directory we are searching */
|
|
struct vnode *tdp; /* saved dp */
|
|
struct mount *mp; /* mount table entry */
|
|
int docache; /* == 0 do not cache last component */
|
|
int wantparent; /* 1 => wantparent or lockparent flag */
|
|
int rdonly; /* lookup read-only flag bit */
|
|
int trailing_slash;
|
|
int error = 0;
|
|
int dpunlocked = 0; /* dp has already been unlocked */
|
|
struct componentname *cnp = &ndp->ni_cnd;
|
|
struct thread *td = cnp->cn_thread;
|
|
int vfslocked;
|
|
int tvfslocked;
|
|
|
|
/*
|
|
* Setup: break out flag bits into variables.
|
|
*/
|
|
vfslocked = (ndp->ni_cnd.cn_flags & GIANTHELD) != 0;
|
|
ndp->ni_cnd.cn_flags &= ~GIANTHELD;
|
|
wantparent = cnp->cn_flags & (LOCKPARENT | WANTPARENT);
|
|
docache = (cnp->cn_flags & NOCACHE) ^ NOCACHE;
|
|
if (cnp->cn_nameiop == DELETE ||
|
|
(wantparent && cnp->cn_nameiop != CREATE &&
|
|
cnp->cn_nameiop != LOOKUP))
|
|
docache = 0;
|
|
rdonly = cnp->cn_flags & RDONLY;
|
|
ndp->ni_dvp = NULL;
|
|
cnp->cn_flags &= ~ISSYMLINK;
|
|
dp = ndp->ni_startdir;
|
|
ndp->ni_startdir = NULLVP;
|
|
vn_lock(dp, LK_EXCLUSIVE | LK_RETRY, td);
|
|
|
|
dirloop:
|
|
/*
|
|
* Search a new directory.
|
|
*
|
|
* The last component of the filename is left accessible via
|
|
* cnp->cn_nameptr for callers that need the name. Callers needing
|
|
* the name set the SAVENAME flag. When done, they assume
|
|
* responsibility for freeing the pathname buffer.
|
|
*/
|
|
cnp->cn_consume = 0;
|
|
for (cp = cnp->cn_nameptr; *cp != 0 && *cp != '/'; cp++)
|
|
continue;
|
|
cnp->cn_namelen = cp - cnp->cn_nameptr;
|
|
if (cnp->cn_namelen > NAME_MAX) {
|
|
error = ENAMETOOLONG;
|
|
goto bad;
|
|
}
|
|
#ifdef NAMEI_DIAGNOSTIC
|
|
{ char c = *cp;
|
|
*cp = '\0';
|
|
printf("{%s}: ", cnp->cn_nameptr);
|
|
*cp = c; }
|
|
#endif
|
|
ndp->ni_pathlen -= cnp->cn_namelen;
|
|
ndp->ni_next = cp;
|
|
|
|
/*
|
|
* Replace multiple slashes by a single slash and trailing slashes
|
|
* by a null. This must be done before VOP_LOOKUP() because some
|
|
* fs's don't know about trailing slashes. Remember if there were
|
|
* trailing slashes to handle symlinks, existing non-directories
|
|
* and non-existing files that won't be directories specially later.
|
|
*/
|
|
trailing_slash = 0;
|
|
while (*cp == '/' && (cp[1] == '/' || cp[1] == '\0')) {
|
|
cp++;
|
|
ndp->ni_pathlen--;
|
|
if (*cp == '\0') {
|
|
trailing_slash = 1;
|
|
*ndp->ni_next = '\0'; /* XXX for direnter() ... */
|
|
}
|
|
}
|
|
ndp->ni_next = cp;
|
|
|
|
cnp->cn_flags |= MAKEENTRY;
|
|
if (*cp == '\0' && docache == 0)
|
|
cnp->cn_flags &= ~MAKEENTRY;
|
|
if (cnp->cn_namelen == 2 &&
|
|
cnp->cn_nameptr[1] == '.' && cnp->cn_nameptr[0] == '.')
|
|
cnp->cn_flags |= ISDOTDOT;
|
|
else
|
|
cnp->cn_flags &= ~ISDOTDOT;
|
|
if (*ndp->ni_next == 0)
|
|
cnp->cn_flags |= ISLASTCN;
|
|
else
|
|
cnp->cn_flags &= ~ISLASTCN;
|
|
|
|
|
|
/*
|
|
* Check for degenerate name (e.g. / or "")
|
|
* which is a way of talking about a directory,
|
|
* e.g. like "/." or ".".
|
|
*/
|
|
if (cnp->cn_nameptr[0] == '\0') {
|
|
if (dp->v_type != VDIR) {
|
|
error = ENOTDIR;
|
|
goto bad;
|
|
}
|
|
if (cnp->cn_nameiop != LOOKUP) {
|
|
error = EISDIR;
|
|
goto bad;
|
|
}
|
|
if (wantparent) {
|
|
ndp->ni_dvp = dp;
|
|
VREF(dp);
|
|
}
|
|
ndp->ni_vp = dp;
|
|
if (!(cnp->cn_flags & (LOCKPARENT | LOCKLEAF)))
|
|
VOP_UNLOCK(dp, 0, td);
|
|
/* XXX This should probably move to the top of function. */
|
|
if (cnp->cn_flags & SAVESTART)
|
|
panic("lookup: SAVESTART");
|
|
goto success;
|
|
}
|
|
|
|
/*
|
|
* Handle "..": two special cases.
|
|
* 1. If at root directory (e.g. after chroot)
|
|
* or at absolute root directory
|
|
* then ignore it so can't get out.
|
|
* 2. If this vnode is the root of a mounted
|
|
* filesystem, then replace it with the
|
|
* vnode which was mounted on so we take the
|
|
* .. in the other filesystem.
|
|
* 3. If the vnode is the top directory of
|
|
* the jail or chroot, don't let them out.
|
|
*/
|
|
if (cnp->cn_flags & ISDOTDOT) {
|
|
for (;;) {
|
|
if (dp == ndp->ni_rootdir ||
|
|
dp == ndp->ni_topdir ||
|
|
dp == rootvnode) {
|
|
ndp->ni_dvp = dp;
|
|
ndp->ni_vp = dp;
|
|
VREF(dp);
|
|
goto nextname;
|
|
}
|
|
if ((dp->v_vflag & VV_ROOT) == 0 ||
|
|
(cnp->cn_flags & NOCROSSMOUNT))
|
|
break;
|
|
if (dp->v_mount == NULL) { /* forced unmount */
|
|
error = EBADF;
|
|
goto bad;
|
|
}
|
|
tdp = dp;
|
|
tvfslocked = vfslocked;
|
|
dp = dp->v_mount->mnt_vnodecovered;
|
|
vput(tdp);
|
|
vfslocked = VFS_LOCK_GIANT(dp->v_mount);
|
|
VFS_UNLOCK_GIANT(tvfslocked);
|
|
VREF(dp);
|
|
vn_lock(dp, LK_EXCLUSIVE | LK_RETRY, td);
|
|
}
|
|
}
|
|
|
|
/*
|
|
* We now have a segment name to search for, and a directory to search.
|
|
*/
|
|
unionlookup:
|
|
#ifdef MAC
|
|
if ((cnp->cn_flags & NOMACCHECK) == 0) {
|
|
error = mac_check_vnode_lookup(td->td_ucred, dp, cnp);
|
|
if (error)
|
|
goto bad;
|
|
}
|
|
#endif
|
|
ndp->ni_dvp = dp;
|
|
ndp->ni_vp = NULL;
|
|
cnp->cn_flags &= ~PDIRUNLOCK;
|
|
ASSERT_VOP_LOCKED(dp, "lookup");
|
|
#ifdef NAMEI_DIAGNOSTIC
|
|
vprint("lookup in", dp);
|
|
#endif
|
|
if ((error = VOP_LOOKUP(dp, &ndp->ni_vp, cnp)) != 0) {
|
|
KASSERT(ndp->ni_vp == NULL, ("leaf should be empty"));
|
|
#ifdef NAMEI_DIAGNOSTIC
|
|
printf("not found\n");
|
|
#endif
|
|
if ((error == ENOENT) &&
|
|
(dp->v_vflag & VV_ROOT) && (dp->v_mount != NULL) &&
|
|
(dp->v_mount->mnt_flag & MNT_UNION)) {
|
|
tdp = dp;
|
|
tvfslocked = vfslocked;
|
|
dp = dp->v_mount->mnt_vnodecovered;
|
|
if (cnp->cn_flags & PDIRUNLOCK)
|
|
vrele(tdp);
|
|
else
|
|
vput(tdp);
|
|
vfslocked = VFS_LOCK_GIANT(dp->v_mount);
|
|
VFS_UNLOCK_GIANT(tvfslocked);
|
|
VREF(dp);
|
|
vn_lock(dp, LK_EXCLUSIVE | LK_RETRY, td);
|
|
goto unionlookup;
|
|
}
|
|
|
|
if (error != EJUSTRETURN)
|
|
goto bad;
|
|
/*
|
|
* If creating and at end of pathname, then can consider
|
|
* allowing file to be created.
|
|
*/
|
|
if (rdonly) {
|
|
error = EROFS;
|
|
goto bad;
|
|
}
|
|
if (*cp == '\0' && trailing_slash &&
|
|
!(cnp->cn_flags & WILLBEDIR)) {
|
|
error = ENOENT;
|
|
goto bad;
|
|
}
|
|
/*
|
|
* We return with ni_vp NULL to indicate that the entry
|
|
* doesn't currently exist, leaving a pointer to the
|
|
* (possibly locked) directory inode in ndp->ni_dvp.
|
|
*/
|
|
if (cnp->cn_flags & SAVESTART) {
|
|
ndp->ni_startdir = ndp->ni_dvp;
|
|
VREF(ndp->ni_startdir);
|
|
}
|
|
goto success;
|
|
}
|
|
#ifdef NAMEI_DIAGNOSTIC
|
|
printf("found\n");
|
|
#endif
|
|
|
|
ASSERT_VOP_LOCKED(ndp->ni_vp, "lookup");
|
|
|
|
/*
|
|
* Take into account any additional components consumed by
|
|
* the underlying filesystem.
|
|
*/
|
|
if (cnp->cn_consume > 0) {
|
|
cnp->cn_nameptr += cnp->cn_consume;
|
|
ndp->ni_next += cnp->cn_consume;
|
|
ndp->ni_pathlen -= cnp->cn_consume;
|
|
cnp->cn_consume = 0;
|
|
}
|
|
|
|
dp = ndp->ni_vp;
|
|
|
|
/*
|
|
* Check to see if the vnode has been mounted on;
|
|
* if so find the root of the mounted filesystem.
|
|
*/
|
|
while (dp->v_type == VDIR && (mp = dp->v_mountedhere) &&
|
|
(cnp->cn_flags & NOCROSSMOUNT) == 0) {
|
|
if (vfs_busy(mp, 0, 0, td))
|
|
continue;
|
|
VOP_UNLOCK(dp, 0, td);
|
|
tvfslocked = VFS_LOCK_GIANT(mp);
|
|
error = VFS_ROOT(mp, &tdp, td);
|
|
vfs_unbusy(mp, td);
|
|
if (error) {
|
|
VFS_UNLOCK_GIANT(tvfslocked);
|
|
dpunlocked = 1;
|
|
goto bad2;
|
|
}
|
|
vrele(dp);
|
|
VFS_UNLOCK_GIANT(vfslocked);
|
|
ndp->ni_vp = dp = tdp;
|
|
vfslocked = tvfslocked;
|
|
}
|
|
|
|
/*
|
|
* Check for symbolic link
|
|
*/
|
|
if ((dp->v_type == VLNK) &&
|
|
((cnp->cn_flags & FOLLOW) || trailing_slash ||
|
|
*ndp->ni_next == '/')) {
|
|
cnp->cn_flags |= ISSYMLINK;
|
|
if (dp->v_mount == NULL) {
|
|
/* We can't know whether the directory was mounted with
|
|
* NOSYMFOLLOW, so we can't follow safely. */
|
|
error = EBADF;
|
|
goto bad2;
|
|
}
|
|
if (dp->v_mount->mnt_flag & MNT_NOSYMFOLLOW) {
|
|
error = EACCES;
|
|
goto bad2;
|
|
}
|
|
goto success;
|
|
}
|
|
|
|
/*
|
|
* Check for bogus trailing slashes.
|
|
*/
|
|
if (trailing_slash && dp->v_type != VDIR) {
|
|
error = ENOTDIR;
|
|
goto bad2;
|
|
}
|
|
|
|
nextname:
|
|
/*
|
|
* Not a symbolic link. If more pathname,
|
|
* continue at next component, else return.
|
|
*/
|
|
if (*ndp->ni_next == '/') {
|
|
cnp->cn_nameptr = ndp->ni_next;
|
|
while (*cnp->cn_nameptr == '/') {
|
|
cnp->cn_nameptr++;
|
|
ndp->ni_pathlen--;
|
|
}
|
|
if (ndp->ni_dvp != ndp->ni_vp)
|
|
ASSERT_VOP_UNLOCKED(ndp->ni_dvp, "lookup");
|
|
vrele(ndp->ni_dvp);
|
|
goto dirloop;
|
|
}
|
|
/*
|
|
* Disallow directory write attempts on read-only filesystems.
|
|
*/
|
|
if (rdonly &&
|
|
(cnp->cn_nameiop == DELETE || cnp->cn_nameiop == RENAME)) {
|
|
error = EROFS;
|
|
goto bad2;
|
|
}
|
|
if (cnp->cn_flags & SAVESTART) {
|
|
ndp->ni_startdir = ndp->ni_dvp;
|
|
VREF(ndp->ni_startdir);
|
|
}
|
|
if (!wantparent)
|
|
vrele(ndp->ni_dvp);
|
|
|
|
if ((cnp->cn_flags & LOCKLEAF) == 0)
|
|
VOP_UNLOCK(dp, 0, td);
|
|
success:
|
|
if (vfslocked)
|
|
ndp->ni_cnd.cn_flags |= GIANTHELD;
|
|
return (0);
|
|
|
|
bad2:
|
|
if ((cnp->cn_flags & (LOCKPARENT | PDIRUNLOCK)) == LOCKPARENT &&
|
|
*ndp->ni_next == '\0')
|
|
VOP_UNLOCK(ndp->ni_dvp, 0, td);
|
|
vrele(ndp->ni_dvp);
|
|
bad:
|
|
if (dpunlocked)
|
|
vrele(dp);
|
|
else
|
|
vput(dp);
|
|
VFS_UNLOCK_GIANT(vfslocked);
|
|
ndp->ni_cnd.cn_flags &= ~GIANTHELD;
|
|
ndp->ni_vp = NULL;
|
|
return (error);
|
|
}
|
|
|
|
/*
|
|
* relookup - lookup a path name component
|
|
* Used by lookup to re-aquire things.
|
|
*/
|
|
int
|
|
relookup(dvp, vpp, cnp)
|
|
struct vnode *dvp, **vpp;
|
|
struct componentname *cnp;
|
|
{
|
|
struct thread *td = cnp->cn_thread;
|
|
struct vnode *dp = 0; /* the directory we are searching */
|
|
int docache; /* == 0 do not cache last component */
|
|
int wantparent; /* 1 => wantparent or lockparent flag */
|
|
int rdonly; /* lookup read-only flag bit */
|
|
int error = 0;
|
|
|
|
/*
|
|
* Setup: break out flag bits into variables.
|
|
*/
|
|
wantparent = cnp->cn_flags & (LOCKPARENT|WANTPARENT);
|
|
docache = (cnp->cn_flags & NOCACHE) ^ NOCACHE;
|
|
if (cnp->cn_nameiop == DELETE ||
|
|
(wantparent && cnp->cn_nameiop != CREATE))
|
|
docache = 0;
|
|
rdonly = cnp->cn_flags & RDONLY;
|
|
cnp->cn_flags &= ~ISSYMLINK;
|
|
dp = dvp;
|
|
vn_lock(dp, LK_EXCLUSIVE | LK_RETRY, td);
|
|
|
|
/* dirloop: */
|
|
/*
|
|
* Search a new directory.
|
|
*
|
|
* The last component of the filename is left accessible via
|
|
* cnp->cn_nameptr for callers that need the name. Callers needing
|
|
* the name set the SAVENAME flag. When done, they assume
|
|
* responsibility for freeing the pathname buffer.
|
|
*/
|
|
#ifdef NAMEI_DIAGNOSTIC
|
|
printf("{%s}: ", cnp->cn_nameptr);
|
|
#endif
|
|
|
|
/*
|
|
* Check for degenerate name (e.g. / or "")
|
|
* which is a way of talking about a directory,
|
|
* e.g. like "/." or ".".
|
|
*/
|
|
if (cnp->cn_nameptr[0] == '\0') {
|
|
if (cnp->cn_nameiop != LOOKUP || wantparent) {
|
|
error = EISDIR;
|
|
goto bad;
|
|
}
|
|
if (dp->v_type != VDIR) {
|
|
error = ENOTDIR;
|
|
goto bad;
|
|
}
|
|
if (!(cnp->cn_flags & LOCKLEAF))
|
|
VOP_UNLOCK(dp, 0, td);
|
|
*vpp = dp;
|
|
/* XXX This should probably move to the top of function. */
|
|
if (cnp->cn_flags & SAVESTART)
|
|
panic("lookup: SAVESTART");
|
|
return (0);
|
|
}
|
|
|
|
if (cnp->cn_flags & ISDOTDOT)
|
|
panic ("relookup: lookup on dot-dot");
|
|
|
|
/*
|
|
* We now have a segment name to search for, and a directory to search.
|
|
*/
|
|
#ifdef NAMEI_DIAGNOSTIC
|
|
vprint("search in:", dp);
|
|
#endif
|
|
if ((error = VOP_LOOKUP(dp, vpp, cnp)) != 0) {
|
|
KASSERT(*vpp == NULL, ("leaf should be empty"));
|
|
if (error != EJUSTRETURN)
|
|
goto bad;
|
|
/*
|
|
* If creating and at end of pathname, then can consider
|
|
* allowing file to be created.
|
|
*/
|
|
if (rdonly) {
|
|
error = EROFS;
|
|
goto bad;
|
|
}
|
|
/* ASSERT(dvp == ndp->ni_startdir) */
|
|
if (cnp->cn_flags & SAVESTART)
|
|
VREF(dvp);
|
|
/*
|
|
* We return with ni_vp NULL to indicate that the entry
|
|
* doesn't currently exist, leaving a pointer to the
|
|
* (possibly locked) directory inode in ndp->ni_dvp.
|
|
*/
|
|
return (0);
|
|
}
|
|
dp = *vpp;
|
|
|
|
/*
|
|
* Check for symbolic link
|
|
*/
|
|
KASSERT(dp->v_type != VLNK || !(cnp->cn_flags & FOLLOW),
|
|
("relookup: symlink found.\n"));
|
|
|
|
/*
|
|
* Disallow directory write attempts on read-only filesystems.
|
|
*/
|
|
if (rdonly &&
|
|
(cnp->cn_nameiop == DELETE || cnp->cn_nameiop == RENAME)) {
|
|
error = EROFS;
|
|
goto bad2;
|
|
}
|
|
/* ASSERT(dvp == ndp->ni_startdir) */
|
|
if (cnp->cn_flags & SAVESTART)
|
|
VREF(dvp);
|
|
|
|
if (!wantparent)
|
|
vrele(dvp);
|
|
|
|
if ((cnp->cn_flags & LOCKLEAF) == 0)
|
|
VOP_UNLOCK(dp, 0, td);
|
|
return (0);
|
|
|
|
bad2:
|
|
if ((cnp->cn_flags & LOCKPARENT) && (cnp->cn_flags & ISLASTCN))
|
|
VOP_UNLOCK(dvp, 0, td);
|
|
vrele(dvp);
|
|
bad:
|
|
vput(dp);
|
|
*vpp = NULL;
|
|
return (error);
|
|
}
|
|
|
|
/*
|
|
* Determine if there is a suitable alternate filename under the specified
|
|
* prefix for the specified path. If the create flag is set, then the
|
|
* alternate prefix will be used so long as the parent directory exists.
|
|
* This is used by the various compatiblity ABIs so that Linux binaries prefer
|
|
* files under /compat/linux for example. The chosen path (whether under
|
|
* the prefix or under /) is returned in a kernel malloc'd buffer pointed
|
|
* to by pathbuf. The caller is responsible for free'ing the buffer from
|
|
* the M_TEMP bucket if one is returned.
|
|
*/
|
|
int
|
|
kern_alternate_path(struct thread *td, const char *prefix, char *path,
|
|
enum uio_seg pathseg, char **pathbuf, int create)
|
|
{
|
|
struct nameidata nd, ndroot;
|
|
char *ptr, *buf, *cp;
|
|
size_t len, sz;
|
|
int error;
|
|
|
|
buf = (char *) malloc(MAXPATHLEN, M_TEMP, M_WAITOK);
|
|
*pathbuf = buf;
|
|
|
|
/* Copy the prefix into the new pathname as a starting point. */
|
|
len = strlcpy(buf, prefix, MAXPATHLEN);
|
|
if (len >= MAXPATHLEN) {
|
|
*pathbuf = NULL;
|
|
free(buf, M_TEMP);
|
|
return (EINVAL);
|
|
}
|
|
sz = MAXPATHLEN - len;
|
|
ptr = buf + len;
|
|
|
|
/* Append the filename to the prefix. */
|
|
if (pathseg == UIO_SYSSPACE)
|
|
error = copystr(path, ptr, sz, &len);
|
|
else
|
|
error = copyinstr(path, ptr, sz, &len);
|
|
|
|
if (error) {
|
|
*pathbuf = NULL;
|
|
free(buf, M_TEMP);
|
|
return (error);
|
|
}
|
|
|
|
/* Only use a prefix with absolute pathnames. */
|
|
if (*ptr != '/') {
|
|
error = EINVAL;
|
|
goto keeporig;
|
|
}
|
|
|
|
/* XXX: VFS_LOCK_GIANT? */
|
|
mtx_lock(&Giant);
|
|
|
|
/*
|
|
* We know that there is a / somewhere in this pathname.
|
|
* Search backwards for it, to find the file's parent dir
|
|
* to see if it exists in the alternate tree. If it does,
|
|
* and we want to create a file (cflag is set). We don't
|
|
* need to worry about the root comparison in this case.
|
|
*/
|
|
|
|
if (create) {
|
|
for (cp = &ptr[len] - 1; *cp != '/'; cp--);
|
|
*cp = '\0';
|
|
|
|
NDINIT(&nd, LOOKUP, FOLLOW, UIO_SYSSPACE, buf, td);
|
|
error = namei(&nd);
|
|
*cp = '/';
|
|
if (error != 0)
|
|
goto nd_failed;
|
|
} else {
|
|
NDINIT(&nd, LOOKUP, FOLLOW, UIO_SYSSPACE, buf, td);
|
|
|
|
error = namei(&nd);
|
|
if (error != 0)
|
|
goto nd_failed;
|
|
|
|
/*
|
|
* We now compare the vnode of the prefix to the one
|
|
* vnode asked. If they resolve to be the same, then we
|
|
* ignore the match so that the real root gets used.
|
|
* This avoids the problem of traversing "../.." to find the
|
|
* root directory and never finding it, because "/" resolves
|
|
* to the emulation root directory. This is expensive :-(
|
|
*/
|
|
NDINIT(&ndroot, LOOKUP, FOLLOW, UIO_SYSSPACE, prefix, td);
|
|
|
|
/* We shouldn't ever get an error from this namei(). */
|
|
error = namei(&ndroot);
|
|
if (error == 0) {
|
|
if (nd.ni_vp == ndroot.ni_vp)
|
|
error = ENOENT;
|
|
|
|
NDFREE(&ndroot, NDF_ONLY_PNBUF);
|
|
vrele(ndroot.ni_vp);
|
|
}
|
|
}
|
|
|
|
NDFREE(&nd, NDF_ONLY_PNBUF);
|
|
vrele(nd.ni_vp);
|
|
|
|
nd_failed:
|
|
/* XXX: VFS_UNLOCK_GIANT? */
|
|
mtx_unlock(&Giant);
|
|
|
|
keeporig:
|
|
/* If there was an error, use the original path name. */
|
|
if (error)
|
|
bcopy(ptr, buf, len);
|
|
return (error);
|
|
}
|