mirror of
https://git.FreeBSD.org/src.git
synced 2024-12-25 11:37:56 +00:00
48fbb6cc7e
bad, I will get to it eventually, unless somebody beats me to it.
1995 lines
47 KiB
C
1995 lines
47 KiB
C
/*
|
|
* Copyright (c) 1989, 1993
|
|
* The Regents of the University of California. All rights reserved.
|
|
*
|
|
* This code is derived from software contributed to Berkeley by
|
|
* Rick Macklem at The University of Guelph.
|
|
*
|
|
* Redistribution and use in source and binary forms, with or without
|
|
* modification, are permitted provided that the following conditions
|
|
* are met:
|
|
* 1. Redistributions of source code must retain the above copyright
|
|
* notice, this list of conditions and the following disclaimer.
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
* documentation and/or other materials provided with the distribution.
|
|
* 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.
|
|
*
|
|
* @(#)nfs_serv.c 8.3 (Berkeley) 1/12/94
|
|
* $Id: nfs_serv.c,v 1.6 1994/09/28 16:45:18 dfr Exp $
|
|
*/
|
|
|
|
/*
|
|
* nfs version 2 server calls to vnode ops
|
|
* - these routines generally have 3 phases
|
|
* 1 - break down and validate rpc request in mbuf list
|
|
* 2 - do the vnode ops for the request
|
|
* (surprisingly ?? many are very similar to syscalls in vfs_syscalls.c)
|
|
* 3 - build the rpc reply in an mbuf list
|
|
* nb:
|
|
* - do not mix the phases, since the nfsm_?? macros can return failures
|
|
* on a bad rpc or similar and do not do any vrele() or vput()'s
|
|
*
|
|
* - the nfsm_reply() macro generates an nfs rpc reply with the nfs
|
|
* error number iff error != 0 whereas
|
|
* returning an error from the server function implies a fatal error
|
|
* such as a badly constructed rpc request that should be dropped without
|
|
* a reply.
|
|
*/
|
|
|
|
#include <sys/param.h>
|
|
#include <sys/systm.h>
|
|
#include <sys/proc.h>
|
|
#include <sys/file.h>
|
|
#include <sys/namei.h>
|
|
#include <sys/vnode.h>
|
|
#include <sys/mount.h>
|
|
#include <sys/mbuf.h>
|
|
#include <sys/dirent.h>
|
|
#include <sys/stat.h>
|
|
|
|
#include <vm/vm.h>
|
|
|
|
#include <nfs/nfsv2.h>
|
|
#include <nfs/rpcv2.h>
|
|
#include <nfs/nfs.h>
|
|
#include <nfs/xdr_subs.h>
|
|
#include <nfs/nfsm_subs.h>
|
|
#include <nfs/nqnfs.h>
|
|
|
|
/* Defs */
|
|
#define TRUE 1
|
|
#define FALSE 0
|
|
|
|
/* Global vars */
|
|
extern u_long nfs_procids[NFS_NPROCS];
|
|
extern u_long nfs_xdrneg1;
|
|
extern u_long nfs_false, nfs_true;
|
|
nfstype nfs_type[9] = { NFNON, NFREG, NFDIR, NFBLK, NFCHR, NFLNK, NFNON,
|
|
NFCHR, NFNON };
|
|
|
|
/*
|
|
* nqnfs access service
|
|
*/
|
|
int
|
|
nqnfsrv_access(nfsd, mrep, md, dpos, cred, nam, mrq)
|
|
struct nfsd *nfsd;
|
|
struct mbuf *mrep, *md;
|
|
caddr_t dpos;
|
|
struct ucred *cred;
|
|
struct mbuf *nam, **mrq;
|
|
{
|
|
struct vnode *vp;
|
|
nfsv2fh_t nfh;
|
|
fhandle_t *fhp;
|
|
register u_long *tl;
|
|
register long t1;
|
|
caddr_t bpos;
|
|
int error = 0, rdonly, cache = 0, mode = 0;
|
|
char *cp2;
|
|
struct mbuf *mb, *mreq;
|
|
u_quad_t frev;
|
|
|
|
fhp = &nfh.fh_generic;
|
|
nfsm_srvmtofh(fhp);
|
|
nfsm_dissect(tl, u_long *, 3 * NFSX_UNSIGNED);
|
|
error = nfsrv_fhtovp(fhp, TRUE, &vp, cred, nfsd->nd_slp, nam, &rdonly);
|
|
if (error)
|
|
nfsm_reply(0);
|
|
if (*tl++ == nfs_true)
|
|
mode |= VREAD;
|
|
if (*tl++ == nfs_true)
|
|
mode |= VWRITE;
|
|
if (*tl == nfs_true)
|
|
mode |= VEXEC;
|
|
error = nfsrv_access(vp, mode, cred, rdonly, nfsd->nd_procp);
|
|
vput(vp);
|
|
nfsm_reply(0);
|
|
nfsm_srvdone;
|
|
}
|
|
|
|
/*
|
|
* nfs getattr service
|
|
*/
|
|
int
|
|
nfsrv_getattr(nfsd, mrep, md, dpos, cred, nam, mrq)
|
|
struct nfsd *nfsd;
|
|
struct mbuf *mrep, *md;
|
|
caddr_t dpos;
|
|
struct ucred *cred;
|
|
struct mbuf *nam, **mrq;
|
|
{
|
|
register struct nfsv2_fattr *fp;
|
|
struct vattr va;
|
|
register struct vattr *vap = &va;
|
|
struct vnode *vp;
|
|
nfsv2fh_t nfh;
|
|
fhandle_t *fhp;
|
|
register u_long *tl;
|
|
register long t1;
|
|
caddr_t bpos;
|
|
int error = 0, rdonly, cache;
|
|
char *cp2;
|
|
struct mbuf *mb, *mb2, *mreq;
|
|
u_quad_t frev;
|
|
|
|
fhp = &nfh.fh_generic;
|
|
nfsm_srvmtofh(fhp);
|
|
error = nfsrv_fhtovp(fhp, TRUE, &vp, cred, nfsd->nd_slp, nam, &rdonly);
|
|
if (error)
|
|
nfsm_reply(0);
|
|
nqsrv_getl(vp, NQL_READ);
|
|
error = VOP_GETATTR(vp, vap, cred, nfsd->nd_procp);
|
|
vput(vp);
|
|
nfsm_reply(NFSX_FATTR(nfsd->nd_nqlflag != NQL_NOVAL));
|
|
nfsm_build(fp, struct nfsv2_fattr *, NFSX_FATTR(nfsd->nd_nqlflag != NQL_NOVAL));
|
|
nfsm_srvfillattr;
|
|
nfsm_srvdone;
|
|
}
|
|
|
|
/*
|
|
* nfs setattr service
|
|
*/
|
|
int
|
|
nfsrv_setattr(nfsd, mrep, md, dpos, cred, nam, mrq)
|
|
struct nfsd *nfsd;
|
|
struct mbuf *mrep, *md;
|
|
caddr_t dpos;
|
|
struct ucred *cred;
|
|
struct mbuf *nam, **mrq;
|
|
{
|
|
struct vattr va;
|
|
register struct vattr *vap = &va;
|
|
register struct nfsv2_sattr *sp;
|
|
register struct nfsv2_fattr *fp;
|
|
struct vnode *vp;
|
|
nfsv2fh_t nfh;
|
|
fhandle_t *fhp;
|
|
register u_long *tl;
|
|
register long t1;
|
|
caddr_t bpos;
|
|
int error = 0, rdonly, cache;
|
|
char *cp2;
|
|
struct mbuf *mb, *mb2, *mreq;
|
|
u_quad_t frev, frev2;
|
|
|
|
fhp = &nfh.fh_generic;
|
|
nfsm_srvmtofh(fhp);
|
|
nfsm_dissect(sp, struct nfsv2_sattr *, NFSX_SATTR(nfsd->nd_nqlflag != NQL_NOVAL));
|
|
error = nfsrv_fhtovp(fhp, TRUE, &vp, cred, nfsd->nd_slp, nam, &rdonly);
|
|
if (error)
|
|
nfsm_reply(0);
|
|
nqsrv_getl(vp, NQL_WRITE);
|
|
VATTR_NULL(vap);
|
|
/*
|
|
* Nah nah nah nah na nah
|
|
* There is a bug in the Sun client that puts 0xffff in the mode
|
|
* field of sattr when it should put in 0xffffffff. The u_short
|
|
* doesn't sign extend.
|
|
* --> check the low order 2 bytes for 0xffff
|
|
*/
|
|
if ((fxdr_unsigned(int, sp->sa_mode) & 0xffff) != 0xffff)
|
|
vap->va_mode = nfstov_mode(sp->sa_mode);
|
|
if (sp->sa_uid != nfs_xdrneg1)
|
|
vap->va_uid = fxdr_unsigned(uid_t, sp->sa_uid);
|
|
if (sp->sa_gid != nfs_xdrneg1)
|
|
vap->va_gid = fxdr_unsigned(gid_t, sp->sa_gid);
|
|
if (nfsd->nd_nqlflag == NQL_NOVAL) {
|
|
if (sp->sa_nfssize != nfs_xdrneg1)
|
|
vap->va_size = fxdr_unsigned(u_quad_t, sp->sa_nfssize);
|
|
if (sp->sa_nfsatime.nfs_sec != nfs_xdrneg1) {
|
|
#ifdef notyet
|
|
fxdr_nfstime(&sp->sa_nfsatime, &vap->va_atime);
|
|
#else
|
|
vap->va_atime.ts_sec =
|
|
fxdr_unsigned(long, sp->sa_nfsatime.nfs_sec);
|
|
vap->va_atime.ts_nsec = 0;
|
|
#endif
|
|
}
|
|
if (sp->sa_nfsmtime.nfs_sec != nfs_xdrneg1)
|
|
fxdr_nfstime(&sp->sa_nfsmtime, &vap->va_mtime);
|
|
} else {
|
|
fxdr_hyper(&sp->sa_nqsize, &vap->va_size);
|
|
fxdr_nqtime(&sp->sa_nqatime, &vap->va_atime);
|
|
fxdr_nqtime(&sp->sa_nqmtime, &vap->va_mtime);
|
|
vap->va_flags = fxdr_unsigned(u_long, sp->sa_nqflags);
|
|
}
|
|
|
|
/*
|
|
* If the size is being changed write acces is required, otherwise
|
|
* just check for a read only file system.
|
|
*/
|
|
if (vap->va_size == ((u_quad_t)((quad_t) -1))) {
|
|
if (rdonly || (vp->v_mount->mnt_flag & MNT_RDONLY)) {
|
|
error = EROFS;
|
|
goto out;
|
|
}
|
|
} else {
|
|
if (vp->v_type == VDIR) {
|
|
error = EISDIR;
|
|
goto out;
|
|
} else {
|
|
error = nfsrv_access(vp, VWRITE, cred, rdonly,
|
|
nfsd->nd_procp);
|
|
if (error)
|
|
goto out;
|
|
}
|
|
}
|
|
error = VOP_SETATTR(vp, vap, cred, nfsd->nd_procp);
|
|
if (error) {
|
|
vput(vp);
|
|
nfsm_reply(0);
|
|
}
|
|
error = VOP_GETATTR(vp, vap, cred, nfsd->nd_procp);
|
|
out:
|
|
vput(vp);
|
|
nfsm_reply(NFSX_FATTR(nfsd->nd_nqlflag != NQL_NOVAL) + 2*NFSX_UNSIGNED);
|
|
nfsm_build(fp, struct nfsv2_fattr *, NFSX_FATTR(nfsd->nd_nqlflag != NQL_NOVAL));
|
|
nfsm_srvfillattr;
|
|
if (nfsd->nd_nqlflag != NQL_NOVAL) {
|
|
nfsm_build(tl, u_long *, 2*NFSX_UNSIGNED);
|
|
txdr_hyper(&frev2, tl);
|
|
}
|
|
nfsm_srvdone;
|
|
}
|
|
|
|
/*
|
|
* nfs lookup rpc
|
|
*/
|
|
int
|
|
nfsrv_lookup(nfsd, mrep, md, dpos, cred, nam, mrq)
|
|
struct nfsd *nfsd;
|
|
struct mbuf *mrep, *md;
|
|
caddr_t dpos;
|
|
struct ucred *cred;
|
|
struct mbuf *nam, **mrq;
|
|
{
|
|
register struct nfsv2_fattr *fp;
|
|
struct nameidata nd;
|
|
struct vnode *vp;
|
|
nfsv2fh_t nfh;
|
|
fhandle_t *fhp;
|
|
register caddr_t cp;
|
|
register u_long *tl;
|
|
register long t1;
|
|
caddr_t bpos;
|
|
int error = 0, cache, cache2, len;
|
|
u_long duration2;
|
|
char *cp2;
|
|
struct mbuf *mb, *mb2, *mreq;
|
|
struct vattr va, *vap = &va;
|
|
u_quad_t frev, frev2;
|
|
|
|
fhp = &nfh.fh_generic;
|
|
duration2 = 0;
|
|
if (nfsd->nd_nqlflag != NQL_NOVAL) {
|
|
nfsm_dissect(tl, u_long *, NFSX_UNSIGNED);
|
|
duration2 = fxdr_unsigned(int, *tl);
|
|
}
|
|
nfsm_srvmtofh(fhp);
|
|
nfsm_srvstrsiz(len, NFS_MAXNAMLEN);
|
|
nd.ni_cnd.cn_cred = cred;
|
|
nd.ni_cnd.cn_nameiop = LOOKUP;
|
|
nd.ni_cnd.cn_flags = LOCKLEAF | SAVESTART;
|
|
error = nfs_namei(&nd, fhp, len, nfsd->nd_slp, nam, &md, &dpos,
|
|
nfsd->nd_procp);
|
|
if (error)
|
|
nfsm_reply(0);
|
|
nqsrv_getl(nd.ni_startdir, NQL_READ);
|
|
vrele(nd.ni_startdir);
|
|
FREE(nd.ni_cnd.cn_pnbuf, M_NAMEI);
|
|
vp = nd.ni_vp;
|
|
bzero((caddr_t)fhp, sizeof(nfh));
|
|
fhp->fh_fsid = vp->v_mount->mnt_stat.f_fsid;
|
|
error = VFS_VPTOFH(vp, &fhp->fh_fid);
|
|
if (error) {
|
|
vput(vp);
|
|
nfsm_reply(0);
|
|
}
|
|
if (duration2)
|
|
(void) nqsrv_getlease(vp, &duration2, NQL_READ, nfsd,
|
|
nam, &cache2, &frev2, cred);
|
|
error = VOP_GETATTR(vp, vap, cred, nfsd->nd_procp);
|
|
vput(vp);
|
|
nfsm_reply(NFSX_FH + NFSX_FATTR(nfsd->nd_nqlflag != NQL_NOVAL) + 5*NFSX_UNSIGNED);
|
|
if (nfsd->nd_nqlflag != NQL_NOVAL) {
|
|
if (duration2) {
|
|
nfsm_build(tl, u_long *, 5*NFSX_UNSIGNED);
|
|
*tl++ = txdr_unsigned(NQL_READ);
|
|
*tl++ = txdr_unsigned(cache2);
|
|
*tl++ = txdr_unsigned(duration2);
|
|
txdr_hyper(&frev2, tl);
|
|
} else {
|
|
nfsm_build(tl, u_long *, NFSX_UNSIGNED);
|
|
*tl = 0;
|
|
}
|
|
}
|
|
nfsm_srvfhtom(fhp);
|
|
nfsm_build(fp, struct nfsv2_fattr *, NFSX_FATTR(nfsd->nd_nqlflag != NQL_NOVAL));
|
|
nfsm_srvfillattr;
|
|
nfsm_srvdone;
|
|
}
|
|
|
|
/*
|
|
* nfs readlink service
|
|
*/
|
|
int
|
|
nfsrv_readlink(nfsd, mrep, md, dpos, cred, nam, mrq)
|
|
struct nfsd *nfsd;
|
|
struct mbuf *mrep, *md;
|
|
caddr_t dpos;
|
|
struct ucred *cred;
|
|
struct mbuf *nam, **mrq;
|
|
{
|
|
struct iovec iv[(NFS_MAXPATHLEN+MLEN-1)/MLEN];
|
|
register struct iovec *ivp = iv;
|
|
register struct mbuf *mp;
|
|
register u_long *tl;
|
|
register long t1;
|
|
caddr_t bpos;
|
|
int error = 0, rdonly, cache, i, tlen, len;
|
|
char *cp2;
|
|
struct mbuf *mb, *mb2, *mp2 = 0, *mp3 = 0, *mreq;
|
|
struct vnode *vp;
|
|
nfsv2fh_t nfh;
|
|
fhandle_t *fhp;
|
|
struct uio io, *uiop = &io;
|
|
u_quad_t frev;
|
|
|
|
fhp = &nfh.fh_generic;
|
|
nfsm_srvmtofh(fhp);
|
|
len = 0;
|
|
i = 0;
|
|
while (len < NFS_MAXPATHLEN) {
|
|
MGET(mp, M_WAIT, MT_DATA);
|
|
MCLGET(mp, M_WAIT);
|
|
mp->m_len = NFSMSIZ(mp);
|
|
if (len == 0)
|
|
mp3 = mp2 = mp;
|
|
else {
|
|
mp2->m_next = mp;
|
|
mp2 = mp;
|
|
}
|
|
if ((len+mp->m_len) > NFS_MAXPATHLEN) {
|
|
mp->m_len = NFS_MAXPATHLEN-len;
|
|
len = NFS_MAXPATHLEN;
|
|
} else
|
|
len += mp->m_len;
|
|
ivp->iov_base = mtod(mp, caddr_t);
|
|
ivp->iov_len = mp->m_len;
|
|
i++;
|
|
ivp++;
|
|
}
|
|
uiop->uio_iov = iv;
|
|
uiop->uio_iovcnt = i;
|
|
uiop->uio_offset = 0;
|
|
uiop->uio_resid = len;
|
|
uiop->uio_rw = UIO_READ;
|
|
uiop->uio_segflg = UIO_SYSSPACE;
|
|
uiop->uio_procp = (struct proc *)0;
|
|
error = nfsrv_fhtovp(fhp, TRUE, &vp, cred, nfsd->nd_slp, nam, &rdonly);
|
|
if (error) {
|
|
m_freem(mp3);
|
|
nfsm_reply(0);
|
|
}
|
|
if (vp->v_type != VLNK) {
|
|
error = EINVAL;
|
|
goto out;
|
|
}
|
|
nqsrv_getl(vp, NQL_READ);
|
|
error = VOP_READLINK(vp, uiop, cred);
|
|
out:
|
|
vput(vp);
|
|
if (error)
|
|
m_freem(mp3);
|
|
nfsm_reply(NFSX_UNSIGNED);
|
|
if (uiop->uio_resid > 0) {
|
|
len -= uiop->uio_resid;
|
|
tlen = nfsm_rndup(len);
|
|
nfsm_adj(mp3, NFS_MAXPATHLEN-tlen, tlen-len);
|
|
}
|
|
nfsm_build(tl, u_long *, NFSX_UNSIGNED);
|
|
*tl = txdr_unsigned(len);
|
|
mb->m_next = mp3;
|
|
nfsm_srvdone;
|
|
}
|
|
|
|
/*
|
|
* nfs read service
|
|
*/
|
|
int
|
|
nfsrv_read(nfsd, mrep, md, dpos, cred, nam, mrq)
|
|
struct nfsd *nfsd;
|
|
struct mbuf *mrep, *md;
|
|
caddr_t dpos;
|
|
struct ucred *cred;
|
|
struct mbuf *nam, **mrq;
|
|
{
|
|
register struct iovec *iv;
|
|
struct iovec *iv2;
|
|
register struct mbuf *m;
|
|
register struct nfsv2_fattr *fp;
|
|
register u_long *tl;
|
|
register long t1;
|
|
caddr_t bpos;
|
|
int error = 0, rdonly, cache, i, cnt, len, left, siz, tlen;
|
|
char *cp2;
|
|
struct mbuf *mb, *mb2, *mreq;
|
|
struct mbuf *m2;
|
|
struct vnode *vp;
|
|
nfsv2fh_t nfh;
|
|
fhandle_t *fhp;
|
|
struct uio io, *uiop = &io;
|
|
struct vattr va, *vap = &va;
|
|
off_t off;
|
|
u_quad_t frev;
|
|
|
|
fhp = &nfh.fh_generic;
|
|
nfsm_srvmtofh(fhp);
|
|
if (nfsd->nd_nqlflag == NQL_NOVAL) {
|
|
nfsm_dissect(tl, u_long *, NFSX_UNSIGNED);
|
|
off = (off_t)fxdr_unsigned(u_long, *tl);
|
|
} else {
|
|
nfsm_dissect(tl, u_long *, 2 * NFSX_UNSIGNED);
|
|
fxdr_hyper(tl, &off);
|
|
}
|
|
nfsm_srvstrsiz(cnt, NFS_MAXDATA);
|
|
error = nfsrv_fhtovp(fhp, TRUE, &vp, cred, nfsd->nd_slp, nam, &rdonly);
|
|
if (error)
|
|
nfsm_reply(0);
|
|
if (vp->v_type != VREG) {
|
|
error = (vp->v_type == VDIR) ? EISDIR : EACCES;
|
|
vput(vp);
|
|
nfsm_reply(0);
|
|
}
|
|
nqsrv_getl(vp, NQL_READ);
|
|
if ((error = nfsrv_access(vp, VREAD, cred, rdonly, nfsd->nd_procp)) &&
|
|
(error = nfsrv_access(vp, VEXEC, cred, rdonly, nfsd->nd_procp))) {
|
|
vput(vp);
|
|
nfsm_reply(0);
|
|
}
|
|
error = VOP_GETATTR(vp, vap, cred, nfsd->nd_procp);
|
|
if (error) {
|
|
vput(vp);
|
|
nfsm_reply(0);
|
|
}
|
|
if (off >= vap->va_size)
|
|
cnt = 0;
|
|
else if ((off + cnt) > vap->va_size)
|
|
cnt = nfsm_rndup(vap->va_size - off);
|
|
nfsm_reply(NFSX_FATTR(nfsd->nd_nqlflag != NQL_NOVAL)+NFSX_UNSIGNED+nfsm_rndup(cnt));
|
|
nfsm_build(fp, struct nfsv2_fattr *, NFSX_FATTR(nfsd->nd_nqlflag != NQL_NOVAL));
|
|
nfsm_build(tl, u_long *, NFSX_UNSIGNED);
|
|
len = left = cnt;
|
|
if (cnt > 0) {
|
|
/*
|
|
* Generate the mbuf list with the uio_iov ref. to it.
|
|
*/
|
|
i = 0;
|
|
m = m2 = mb;
|
|
MALLOC(iv, struct iovec *,
|
|
((NFS_MAXDATA+MLEN-1)/MLEN) * sizeof (struct iovec),
|
|
M_TEMP, M_WAITOK);
|
|
iv2 = iv;
|
|
while (left > 0) {
|
|
siz = min(M_TRAILINGSPACE(m), left);
|
|
if (siz > 0) {
|
|
m->m_len += siz;
|
|
iv->iov_base = bpos;
|
|
iv->iov_len = siz;
|
|
iv++;
|
|
i++;
|
|
left -= siz;
|
|
}
|
|
if (left > 0) {
|
|
MGET(m, M_WAIT, MT_DATA);
|
|
MCLGET(m, M_WAIT);
|
|
m->m_len = 0;
|
|
m2->m_next = m;
|
|
m2 = m;
|
|
bpos = mtod(m, caddr_t);
|
|
}
|
|
}
|
|
uiop->uio_iov = iv2;
|
|
uiop->uio_iovcnt = i;
|
|
uiop->uio_offset = off;
|
|
uiop->uio_resid = cnt;
|
|
uiop->uio_rw = UIO_READ;
|
|
uiop->uio_segflg = UIO_SYSSPACE;
|
|
error = VOP_READ(vp, uiop, IO_NODELOCKED, cred);
|
|
off = uiop->uio_offset;
|
|
FREE((caddr_t)iv2, M_TEMP);
|
|
if (error || (error = VOP_GETATTR(vp, vap, cred, nfsd->nd_procp))) {
|
|
m_freem(mreq);
|
|
vput(vp);
|
|
nfsm_reply(0);
|
|
}
|
|
} else
|
|
uiop->uio_resid = 0;
|
|
vput(vp);
|
|
nfsm_srvfillattr;
|
|
len -= uiop->uio_resid;
|
|
tlen = nfsm_rndup(len);
|
|
if (cnt != tlen || tlen != len)
|
|
nfsm_adj(mb, cnt-tlen, tlen-len);
|
|
*tl = txdr_unsigned(len);
|
|
nfsm_srvdone;
|
|
}
|
|
|
|
/*
|
|
* nfs write service
|
|
*/
|
|
int
|
|
nfsrv_write(nfsd, mrep, md, dpos, cred, nam, mrq)
|
|
struct nfsd *nfsd;
|
|
struct mbuf *mrep, *md;
|
|
caddr_t dpos;
|
|
struct ucred *cred;
|
|
struct mbuf *nam, **mrq;
|
|
{
|
|
register struct iovec *ivp;
|
|
register struct mbuf *mp;
|
|
register struct nfsv2_fattr *fp;
|
|
struct iovec iv[NFS_MAXIOVEC];
|
|
struct vattr va;
|
|
register struct vattr *vap = &va;
|
|
register u_long *tl;
|
|
register long t1;
|
|
caddr_t bpos;
|
|
int error = 0, rdonly, cache, siz, len, xfer;
|
|
int ioflags = IO_SYNC | IO_NODELOCKED;
|
|
char *cp2;
|
|
struct mbuf *mb, *mb2, *mreq;
|
|
struct vnode *vp;
|
|
nfsv2fh_t nfh;
|
|
fhandle_t *fhp;
|
|
struct uio io, *uiop = &io;
|
|
off_t off;
|
|
u_quad_t frev;
|
|
|
|
fhp = &nfh.fh_generic;
|
|
nfsm_srvmtofh(fhp);
|
|
nfsm_dissect(tl, u_long *, 4 * NFSX_UNSIGNED);
|
|
if (nfsd->nd_nqlflag == NQL_NOVAL) {
|
|
off = (off_t)fxdr_unsigned(u_long, *++tl);
|
|
tl += 2;
|
|
} else {
|
|
fxdr_hyper(tl, &off);
|
|
tl += 2;
|
|
if (fxdr_unsigned(u_long, *tl++))
|
|
ioflags |= IO_APPEND;
|
|
}
|
|
len = fxdr_unsigned(long, *tl);
|
|
if (len > NFS_MAXDATA || len <= 0) {
|
|
error = EBADRPC;
|
|
nfsm_reply(0);
|
|
}
|
|
if (dpos == (mtod(md, caddr_t)+md->m_len)) {
|
|
mp = md->m_next;
|
|
if (mp == NULL) {
|
|
error = EBADRPC;
|
|
nfsm_reply(0);
|
|
}
|
|
} else {
|
|
mp = md;
|
|
siz = dpos-mtod(mp, caddr_t);
|
|
mp->m_len -= siz;
|
|
NFSMADV(mp, siz);
|
|
}
|
|
error = nfsrv_fhtovp(fhp, TRUE, &vp, cred, nfsd->nd_slp, nam, &rdonly);
|
|
if (error)
|
|
nfsm_reply(0);
|
|
if (vp->v_type != VREG) {
|
|
error = (vp->v_type == VDIR) ? EISDIR : EACCES;
|
|
vput(vp);
|
|
nfsm_reply(0);
|
|
}
|
|
nqsrv_getl(vp, NQL_WRITE);
|
|
error = nfsrv_access(vp, VWRITE, cred, rdonly, nfsd->nd_procp);
|
|
if (error) {
|
|
vput(vp);
|
|
nfsm_reply(0);
|
|
}
|
|
uiop->uio_resid = 0;
|
|
uiop->uio_rw = UIO_WRITE;
|
|
uiop->uio_segflg = UIO_SYSSPACE;
|
|
uiop->uio_procp = (struct proc *)0;
|
|
/*
|
|
* Do up to NFS_MAXIOVEC mbufs of write each iteration of the
|
|
* loop until done.
|
|
*/
|
|
while (len > 0 && uiop->uio_resid == 0) {
|
|
ivp = iv;
|
|
siz = 0;
|
|
uiop->uio_iov = ivp;
|
|
uiop->uio_iovcnt = 0;
|
|
uiop->uio_offset = off;
|
|
while (len > 0 && uiop->uio_iovcnt < NFS_MAXIOVEC && mp != NULL) {
|
|
ivp->iov_base = mtod(mp, caddr_t);
|
|
if (len < mp->m_len)
|
|
ivp->iov_len = xfer = len;
|
|
else
|
|
ivp->iov_len = xfer = mp->m_len;
|
|
#ifdef notdef
|
|
/* Not Yet .. */
|
|
if (M_HASCL(mp) && (((u_long)ivp->iov_base) & CLOFSET) == 0)
|
|
ivp->iov_op = NULL; /* what should it be ?? */
|
|
else
|
|
ivp->iov_op = NULL;
|
|
#endif
|
|
uiop->uio_iovcnt++;
|
|
ivp++;
|
|
len -= xfer;
|
|
siz += xfer;
|
|
mp = mp->m_next;
|
|
}
|
|
if (len > 0 && mp == NULL) {
|
|
error = EBADRPC;
|
|
vput(vp);
|
|
nfsm_reply(0);
|
|
}
|
|
uiop->uio_resid = siz;
|
|
error = VOP_WRITE(vp, uiop, ioflags, cred);
|
|
if (error) {
|
|
vput(vp);
|
|
nfsm_reply(0);
|
|
}
|
|
off = uiop->uio_offset;
|
|
}
|
|
error = VOP_GETATTR(vp, vap, cred, nfsd->nd_procp);
|
|
vput(vp);
|
|
nfsm_reply(NFSX_FATTR(nfsd->nd_nqlflag != NQL_NOVAL));
|
|
nfsm_build(fp, struct nfsv2_fattr *, NFSX_FATTR(nfsd->nd_nqlflag != NQL_NOVAL));
|
|
nfsm_srvfillattr;
|
|
if (nfsd->nd_nqlflag != NQL_NOVAL) {
|
|
nfsm_build(tl, u_long *, 2*NFSX_UNSIGNED);
|
|
txdr_hyper(&vap->va_filerev, tl);
|
|
}
|
|
nfsm_srvdone;
|
|
}
|
|
|
|
/*
|
|
* nfs create service
|
|
* now does a truncate to 0 length via. setattr if it already exists
|
|
*/
|
|
int
|
|
nfsrv_create(nfsd, mrep, md, dpos, cred, nam, mrq)
|
|
struct nfsd *nfsd;
|
|
struct mbuf *mrep, *md;
|
|
caddr_t dpos;
|
|
struct ucred *cred;
|
|
struct mbuf *nam, **mrq;
|
|
{
|
|
register struct nfsv2_fattr *fp;
|
|
struct vattr va;
|
|
register struct vattr *vap = &va;
|
|
register struct nfsv2_sattr *sp;
|
|
register u_long *tl;
|
|
struct nameidata nd;
|
|
register caddr_t cp;
|
|
register long t1;
|
|
caddr_t bpos;
|
|
int error = 0, rdev, cache, len, tsize;
|
|
char *cp2;
|
|
struct mbuf *mb, *mb2, *mreq;
|
|
struct vnode *vp;
|
|
nfsv2fh_t nfh;
|
|
fhandle_t *fhp;
|
|
u_quad_t frev;
|
|
|
|
nd.ni_cnd.cn_nameiop = 0;
|
|
fhp = &nfh.fh_generic;
|
|
nfsm_srvmtofh(fhp);
|
|
nfsm_srvstrsiz(len, NFS_MAXNAMLEN);
|
|
nd.ni_cnd.cn_cred = cred;
|
|
nd.ni_cnd.cn_nameiop = CREATE;
|
|
nd.ni_cnd.cn_flags = LOCKPARENT | LOCKLEAF | SAVESTART;
|
|
error = nfs_namei(&nd, fhp, len, nfsd->nd_slp, nam, &md, &dpos,
|
|
nfsd->nd_procp);
|
|
if (error)
|
|
nfsm_reply(0);
|
|
VATTR_NULL(vap);
|
|
nfsm_dissect(sp, struct nfsv2_sattr *, NFSX_SATTR(nfsd->nd_nqlflag != NQL_NOVAL));
|
|
/*
|
|
* Iff doesn't exist, create it
|
|
* otherwise just truncate to 0 length
|
|
* should I set the mode too ??
|
|
*/
|
|
if (nd.ni_vp == NULL) {
|
|
vap->va_type = IFTOVT(fxdr_unsigned(u_long, sp->sa_mode));
|
|
if (vap->va_type == VNON)
|
|
vap->va_type = VREG;
|
|
vap->va_mode = nfstov_mode(sp->sa_mode);
|
|
if (nfsd->nd_nqlflag == NQL_NOVAL)
|
|
rdev = fxdr_unsigned(long, sp->sa_nfssize);
|
|
else
|
|
rdev = fxdr_unsigned(long, sp->sa_nqrdev);
|
|
if (vap->va_type == VREG || vap->va_type == VSOCK) {
|
|
vrele(nd.ni_startdir);
|
|
nqsrv_getl(nd.ni_dvp, NQL_WRITE);
|
|
error=VOP_CREATE(nd.ni_dvp, &nd.ni_vp, &nd.ni_cnd, vap);
|
|
if (error)
|
|
nfsm_reply(0);
|
|
FREE(nd.ni_cnd.cn_pnbuf, M_NAMEI);
|
|
} else if (vap->va_type == VCHR || vap->va_type == VBLK ||
|
|
vap->va_type == VFIFO) {
|
|
if (vap->va_type == VCHR && rdev == 0xffffffff)
|
|
vap->va_type = VFIFO;
|
|
if (vap->va_type == VFIFO) {
|
|
} else {
|
|
error = suser(cred, (u_short *)0);
|
|
if (error) {
|
|
VOP_ABORTOP(nd.ni_dvp, &nd.ni_cnd);
|
|
vput(nd.ni_dvp);
|
|
goto out;
|
|
} else
|
|
vap->va_rdev = (dev_t)rdev;
|
|
}
|
|
nqsrv_getl(nd.ni_dvp, NQL_WRITE);
|
|
error=VOP_MKNOD(nd.ni_dvp, &nd.ni_vp, &nd.ni_cnd, vap);
|
|
if (error) {
|
|
vrele(nd.ni_startdir);
|
|
nfsm_reply(0);
|
|
}
|
|
nd.ni_cnd.cn_nameiop = LOOKUP;
|
|
nd.ni_cnd.cn_flags &= ~(LOCKPARENT | SAVESTART);
|
|
nd.ni_cnd.cn_proc = nfsd->nd_procp;
|
|
nd.ni_cnd.cn_cred = nfsd->nd_procp->p_ucred;
|
|
error = lookup(&nd);
|
|
if (error) {
|
|
free(nd.ni_cnd.cn_pnbuf, M_NAMEI);
|
|
nfsm_reply(0);
|
|
}
|
|
FREE(nd.ni_cnd.cn_pnbuf, M_NAMEI);
|
|
if (nd.ni_cnd.cn_flags & ISSYMLINK) {
|
|
vrele(nd.ni_dvp);
|
|
vput(nd.ni_vp);
|
|
VOP_ABORTOP(nd.ni_dvp, &nd.ni_cnd);
|
|
error = EINVAL;
|
|
nfsm_reply(0);
|
|
}
|
|
} else {
|
|
VOP_ABORTOP(nd.ni_dvp, &nd.ni_cnd);
|
|
vput(nd.ni_dvp);
|
|
error = ENXIO;
|
|
goto out;
|
|
}
|
|
vp = nd.ni_vp;
|
|
} else {
|
|
vrele(nd.ni_startdir);
|
|
free(nd.ni_cnd.cn_pnbuf, M_NAMEI);
|
|
vp = nd.ni_vp;
|
|
if (nd.ni_dvp == vp)
|
|
vrele(nd.ni_dvp);
|
|
else
|
|
vput(nd.ni_dvp);
|
|
VOP_ABORTOP(nd.ni_dvp, &nd.ni_cnd);
|
|
if (nfsd->nd_nqlflag == NQL_NOVAL) {
|
|
tsize = fxdr_unsigned(long, sp->sa_nfssize);
|
|
if (tsize != -1)
|
|
vap->va_size = (u_quad_t)tsize;
|
|
else
|
|
vap->va_size = -1;
|
|
} else
|
|
fxdr_hyper(&sp->sa_nqsize, &vap->va_size);
|
|
if (vap->va_size != -1) {
|
|
error = nfsrv_access(vp, VWRITE, cred,
|
|
(nd.ni_cnd.cn_flags & RDONLY), nfsd->nd_procp);
|
|
if (error) {
|
|
vput(vp);
|
|
nfsm_reply(0);
|
|
}
|
|
nqsrv_getl(vp, NQL_WRITE);
|
|
error = VOP_SETATTR(vp, vap, cred, nfsd->nd_procp);
|
|
if (error) {
|
|
vput(vp);
|
|
nfsm_reply(0);
|
|
}
|
|
}
|
|
}
|
|
bzero((caddr_t)fhp, sizeof(nfh));
|
|
fhp->fh_fsid = vp->v_mount->mnt_stat.f_fsid;
|
|
error = VFS_VPTOFH(vp, &fhp->fh_fid);
|
|
if (error) {
|
|
vput(vp);
|
|
nfsm_reply(0);
|
|
}
|
|
error = VOP_GETATTR(vp, vap, cred, nfsd->nd_procp);
|
|
vput(vp);
|
|
nfsm_reply(NFSX_FH+NFSX_FATTR(nfsd->nd_nqlflag != NQL_NOVAL));
|
|
nfsm_srvfhtom(fhp);
|
|
nfsm_build(fp, struct nfsv2_fattr *, NFSX_FATTR(nfsd->nd_nqlflag != NQL_NOVAL));
|
|
nfsm_srvfillattr;
|
|
return (error);
|
|
nfsmout:
|
|
if (nd.ni_cnd.cn_nameiop || nd.ni_cnd.cn_flags)
|
|
vrele(nd.ni_startdir);
|
|
VOP_ABORTOP(nd.ni_dvp, &nd.ni_cnd);
|
|
if (nd.ni_dvp == nd.ni_vp)
|
|
vrele(nd.ni_dvp);
|
|
else
|
|
vput(nd.ni_dvp);
|
|
if (nd.ni_vp)
|
|
vput(nd.ni_vp);
|
|
return (error);
|
|
|
|
out:
|
|
vrele(nd.ni_startdir);
|
|
free(nd.ni_cnd.cn_pnbuf, M_NAMEI);
|
|
nfsm_reply(0);
|
|
return (0);
|
|
}
|
|
|
|
/*
|
|
* nfs remove service
|
|
*/
|
|
int
|
|
nfsrv_remove(nfsd, mrep, md, dpos, cred, nam, mrq)
|
|
struct nfsd *nfsd;
|
|
struct mbuf *mrep, *md;
|
|
caddr_t dpos;
|
|
struct ucred *cred;
|
|
struct mbuf *nam, **mrq;
|
|
{
|
|
struct nameidata nd;
|
|
register u_long *tl;
|
|
register long t1;
|
|
caddr_t bpos;
|
|
int error = 0, cache, len;
|
|
char *cp2;
|
|
struct mbuf *mb, *mreq;
|
|
struct vnode *vp;
|
|
nfsv2fh_t nfh;
|
|
fhandle_t *fhp;
|
|
u_quad_t frev;
|
|
|
|
fhp = &nfh.fh_generic;
|
|
nfsm_srvmtofh(fhp);
|
|
nfsm_srvstrsiz(len, NFS_MAXNAMLEN);
|
|
nd.ni_cnd.cn_cred = cred;
|
|
nd.ni_cnd.cn_nameiop = DELETE;
|
|
nd.ni_cnd.cn_flags = LOCKPARENT | LOCKLEAF;
|
|
error = nfs_namei(&nd, fhp, len, nfsd->nd_slp, nam, &md, &dpos,
|
|
nfsd->nd_procp);
|
|
if (error)
|
|
nfsm_reply(0);
|
|
vp = nd.ni_vp;
|
|
if (vp->v_type == VDIR &&
|
|
(error = suser(cred, (u_short *)0)))
|
|
goto out;
|
|
/*
|
|
* The root of a mounted filesystem cannot be deleted.
|
|
*/
|
|
if (vp->v_flag & VROOT) {
|
|
error = EBUSY;
|
|
goto out;
|
|
}
|
|
(void) vnode_pager_uncache(vp);
|
|
out:
|
|
if (!error) {
|
|
nqsrv_getl(nd.ni_dvp, NQL_WRITE);
|
|
nqsrv_getl(vp, NQL_WRITE);
|
|
error = VOP_REMOVE(nd.ni_dvp, nd.ni_vp, &nd.ni_cnd);
|
|
} else {
|
|
VOP_ABORTOP(nd.ni_dvp, &nd.ni_cnd);
|
|
if (nd.ni_dvp == vp)
|
|
vrele(nd.ni_dvp);
|
|
else
|
|
vput(nd.ni_dvp);
|
|
vput(vp);
|
|
}
|
|
nfsm_reply(0);
|
|
nfsm_srvdone;
|
|
}
|
|
|
|
/*
|
|
* nfs rename service
|
|
*/
|
|
int
|
|
nfsrv_rename(nfsd, mrep, md, dpos, cred, nam, mrq)
|
|
struct nfsd *nfsd;
|
|
struct mbuf *mrep, *md;
|
|
caddr_t dpos;
|
|
struct ucred *cred;
|
|
struct mbuf *nam, **mrq;
|
|
{
|
|
register u_long *tl;
|
|
register long t1;
|
|
caddr_t bpos;
|
|
int error = 0, cache, len, len2;
|
|
char *cp2;
|
|
struct mbuf *mb, *mreq;
|
|
struct nameidata fromnd, tond;
|
|
struct vnode *fvp = 0, *tvp, *tdvp;
|
|
nfsv2fh_t fnfh, tnfh;
|
|
fhandle_t *ffhp, *tfhp;
|
|
u_quad_t frev;
|
|
uid_t saved_uid;
|
|
|
|
ffhp = &fnfh.fh_generic;
|
|
tfhp = &tnfh.fh_generic;
|
|
fromnd.ni_cnd.cn_nameiop = 0;
|
|
tond.ni_cnd.cn_nameiop = 0;
|
|
nfsm_srvmtofh(ffhp);
|
|
nfsm_srvstrsiz(len, NFS_MAXNAMLEN);
|
|
/*
|
|
* Remember our original uid so that we can reset cr_uid before
|
|
* the second nfs_namei() call, in case it is remapped.
|
|
*/
|
|
saved_uid = cred->cr_uid;
|
|
fromnd.ni_cnd.cn_cred = cred;
|
|
fromnd.ni_cnd.cn_nameiop = DELETE;
|
|
fromnd.ni_cnd.cn_flags = WANTPARENT | SAVESTART;
|
|
error = nfs_namei(&fromnd, ffhp, len, nfsd->nd_slp, nam, &md, &dpos,
|
|
nfsd->nd_procp);
|
|
if(error)
|
|
nfsm_reply(0);
|
|
fvp = fromnd.ni_vp;
|
|
nfsm_srvmtofh(tfhp);
|
|
nfsm_strsiz(len2, NFS_MAXNAMLEN);
|
|
cred->cr_uid = saved_uid;
|
|
tond.ni_cnd.cn_cred = cred;
|
|
tond.ni_cnd.cn_nameiop = RENAME;
|
|
tond.ni_cnd.cn_flags = LOCKPARENT | LOCKLEAF | NOCACHE | SAVESTART;
|
|
error = nfs_namei(&tond, tfhp, len2, nfsd->nd_slp, nam, &md,
|
|
&dpos, nfsd->nd_procp);
|
|
if (error) {
|
|
VOP_ABORTOP(fromnd.ni_dvp, &fromnd.ni_cnd);
|
|
vrele(fromnd.ni_dvp);
|
|
vrele(fvp);
|
|
goto out1;
|
|
}
|
|
tdvp = tond.ni_dvp;
|
|
tvp = tond.ni_vp;
|
|
if (tvp != NULL) {
|
|
if (fvp->v_type == VDIR && tvp->v_type != VDIR) {
|
|
error = EISDIR;
|
|
goto out;
|
|
} else if (fvp->v_type != VDIR && tvp->v_type == VDIR) {
|
|
error = ENOTDIR;
|
|
goto out;
|
|
}
|
|
if (tvp->v_type == VDIR && tvp->v_mountedhere) {
|
|
error = EXDEV;
|
|
goto out;
|
|
}
|
|
}
|
|
if (fvp->v_type == VDIR && fvp->v_mountedhere) {
|
|
error = EBUSY;
|
|
goto out;
|
|
}
|
|
if (fvp->v_mount != tdvp->v_mount) {
|
|
error = EXDEV;
|
|
goto out;
|
|
}
|
|
if (fvp == tdvp)
|
|
error = EINVAL;
|
|
/*
|
|
* If source is the same as the destination (that is the
|
|
* same vnode with the same name in the same directory),
|
|
* then there is nothing to do.
|
|
*/
|
|
if (fvp == tvp && fromnd.ni_dvp == tdvp &&
|
|
fromnd.ni_cnd.cn_namelen == tond.ni_cnd.cn_namelen &&
|
|
!bcmp(fromnd.ni_cnd.cn_nameptr, tond.ni_cnd.cn_nameptr,
|
|
fromnd.ni_cnd.cn_namelen))
|
|
error = -1;
|
|
out:
|
|
if (!error) {
|
|
nqsrv_getl(fromnd.ni_dvp, NQL_WRITE);
|
|
nqsrv_getl(tdvp, NQL_WRITE);
|
|
if (tvp)
|
|
nqsrv_getl(tvp, NQL_WRITE);
|
|
error = VOP_RENAME(fromnd.ni_dvp, fromnd.ni_vp, &fromnd.ni_cnd,
|
|
tond.ni_dvp, tond.ni_vp, &tond.ni_cnd);
|
|
} else {
|
|
VOP_ABORTOP(tond.ni_dvp, &tond.ni_cnd);
|
|
if (tdvp == tvp)
|
|
vrele(tdvp);
|
|
else
|
|
vput(tdvp);
|
|
if (tvp)
|
|
vput(tvp);
|
|
VOP_ABORTOP(fromnd.ni_dvp, &fromnd.ni_cnd);
|
|
vrele(fromnd.ni_dvp);
|
|
vrele(fvp);
|
|
}
|
|
vrele(tond.ni_startdir);
|
|
FREE(tond.ni_cnd.cn_pnbuf, M_NAMEI);
|
|
out1:
|
|
vrele(fromnd.ni_startdir);
|
|
FREE(fromnd.ni_cnd.cn_pnbuf, M_NAMEI);
|
|
nfsm_reply(0);
|
|
return (error);
|
|
|
|
nfsmout:
|
|
if (tond.ni_cnd.cn_nameiop || tond.ni_cnd.cn_flags) {
|
|
vrele(tond.ni_startdir);
|
|
FREE(tond.ni_cnd.cn_pnbuf, M_NAMEI);
|
|
}
|
|
if (fromnd.ni_cnd.cn_nameiop || fromnd.ni_cnd.cn_flags) {
|
|
vrele(fromnd.ni_startdir);
|
|
FREE(fromnd.ni_cnd.cn_pnbuf, M_NAMEI);
|
|
VOP_ABORTOP(fromnd.ni_dvp, &fromnd.ni_cnd);
|
|
vrele(fromnd.ni_dvp);
|
|
vrele(fvp);
|
|
}
|
|
return (error);
|
|
}
|
|
|
|
/*
|
|
* nfs link service
|
|
*/
|
|
int
|
|
nfsrv_link(nfsd, mrep, md, dpos, cred, nam, mrq)
|
|
struct nfsd *nfsd;
|
|
struct mbuf *mrep, *md;
|
|
caddr_t dpos;
|
|
struct ucred *cred;
|
|
struct mbuf *nam, **mrq;
|
|
{
|
|
struct nameidata nd;
|
|
register u_long *tl;
|
|
register long t1;
|
|
caddr_t bpos;
|
|
int error = 0, rdonly, cache, len;
|
|
char *cp2;
|
|
struct mbuf *mb, *mreq;
|
|
struct vnode *vp, *xp;
|
|
nfsv2fh_t nfh, dnfh;
|
|
fhandle_t *fhp, *dfhp;
|
|
u_quad_t frev;
|
|
|
|
fhp = &nfh.fh_generic;
|
|
dfhp = &dnfh.fh_generic;
|
|
nfsm_srvmtofh(fhp);
|
|
nfsm_srvmtofh(dfhp);
|
|
nfsm_srvstrsiz(len, NFS_MAXNAMLEN);
|
|
error = nfsrv_fhtovp(fhp, FALSE, &vp, cred, nfsd->nd_slp, nam, &rdonly);
|
|
if (error)
|
|
nfsm_reply(0);
|
|
if (vp->v_type == VDIR && (error = suser(cred, (u_short *)0)))
|
|
goto out1;
|
|
nd.ni_cnd.cn_cred = cred;
|
|
nd.ni_cnd.cn_nameiop = CREATE;
|
|
nd.ni_cnd.cn_flags = LOCKPARENT;
|
|
error = nfs_namei(&nd, dfhp, len, nfsd->nd_slp, nam, &md, &dpos,
|
|
nfsd->nd_procp);
|
|
if (error)
|
|
goto out1;
|
|
xp = nd.ni_vp;
|
|
if (xp != NULL) {
|
|
error = EEXIST;
|
|
goto out;
|
|
}
|
|
xp = nd.ni_dvp;
|
|
if (vp->v_mount != xp->v_mount)
|
|
error = EXDEV;
|
|
out:
|
|
if (!error) {
|
|
nqsrv_getl(vp, NQL_WRITE);
|
|
nqsrv_getl(xp, NQL_WRITE);
|
|
error = VOP_LINK(nd.ni_dvp, vp, &nd.ni_cnd);
|
|
} else {
|
|
VOP_ABORTOP(nd.ni_dvp, &nd.ni_cnd);
|
|
if (nd.ni_dvp == nd.ni_vp)
|
|
vrele(nd.ni_dvp);
|
|
else
|
|
vput(nd.ni_dvp);
|
|
if (nd.ni_vp)
|
|
vrele(nd.ni_vp);
|
|
}
|
|
out1:
|
|
vrele(vp);
|
|
nfsm_reply(0);
|
|
nfsm_srvdone;
|
|
}
|
|
|
|
/*
|
|
* nfs symbolic link service
|
|
*/
|
|
int
|
|
nfsrv_symlink(nfsd, mrep, md, dpos, cred, nam, mrq)
|
|
struct nfsd *nfsd;
|
|
struct mbuf *mrep, *md;
|
|
caddr_t dpos;
|
|
struct ucred *cred;
|
|
struct mbuf *nam, **mrq;
|
|
{
|
|
struct vattr va;
|
|
struct nameidata nd;
|
|
register struct vattr *vap = &va;
|
|
register u_long *tl;
|
|
register long t1;
|
|
struct nfsv2_sattr *sp;
|
|
caddr_t bpos;
|
|
struct uio io;
|
|
struct iovec iv;
|
|
int error = 0, cache, len, len2;
|
|
char *pathcp, *cp2;
|
|
struct mbuf *mb, *mreq;
|
|
nfsv2fh_t nfh;
|
|
fhandle_t *fhp;
|
|
u_quad_t frev;
|
|
|
|
pathcp = (char *)0;
|
|
fhp = &nfh.fh_generic;
|
|
nfsm_srvmtofh(fhp);
|
|
nfsm_srvstrsiz(len, NFS_MAXNAMLEN);
|
|
nd.ni_cnd.cn_cred = cred;
|
|
nd.ni_cnd.cn_nameiop = CREATE;
|
|
nd.ni_cnd.cn_flags = LOCKPARENT;
|
|
error = nfs_namei(&nd, fhp, len, nfsd->nd_slp, nam, &md, &dpos,
|
|
nfsd->nd_procp);
|
|
if (error)
|
|
goto out;
|
|
nfsm_strsiz(len2, NFS_MAXPATHLEN);
|
|
MALLOC(pathcp, caddr_t, len2 + 1, M_TEMP, M_WAITOK);
|
|
iv.iov_base = pathcp;
|
|
iv.iov_len = len2;
|
|
io.uio_resid = len2;
|
|
io.uio_offset = 0;
|
|
io.uio_iov = &iv;
|
|
io.uio_iovcnt = 1;
|
|
io.uio_segflg = UIO_SYSSPACE;
|
|
io.uio_rw = UIO_READ;
|
|
io.uio_procp = (struct proc *)0;
|
|
nfsm_mtouio(&io, len2);
|
|
nfsm_dissect(sp, struct nfsv2_sattr *, NFSX_SATTR(nfsd->nd_nqlflag != NQL_NOVAL));
|
|
*(pathcp + len2) = '\0';
|
|
if (nd.ni_vp) {
|
|
VOP_ABORTOP(nd.ni_dvp, &nd.ni_cnd);
|
|
if (nd.ni_dvp == nd.ni_vp)
|
|
vrele(nd.ni_dvp);
|
|
else
|
|
vput(nd.ni_dvp);
|
|
vrele(nd.ni_vp);
|
|
error = EEXIST;
|
|
goto out;
|
|
}
|
|
VATTR_NULL(vap);
|
|
vap->va_mode = fxdr_unsigned(u_short, sp->sa_mode);
|
|
nqsrv_getl(nd.ni_dvp, NQL_WRITE);
|
|
error = VOP_SYMLINK(nd.ni_dvp, &nd.ni_vp, &nd.ni_cnd, vap, pathcp);
|
|
out:
|
|
if (pathcp)
|
|
FREE(pathcp, M_TEMP);
|
|
nfsm_reply(0);
|
|
return (error);
|
|
nfsmout:
|
|
VOP_ABORTOP(nd.ni_dvp, &nd.ni_cnd);
|
|
if (nd.ni_dvp == nd.ni_vp)
|
|
vrele(nd.ni_dvp);
|
|
else
|
|
vput(nd.ni_dvp);
|
|
if (nd.ni_vp)
|
|
vrele(nd.ni_vp);
|
|
if (pathcp)
|
|
FREE(pathcp, M_TEMP);
|
|
return (error);
|
|
}
|
|
|
|
/*
|
|
* nfs mkdir service
|
|
*/
|
|
int
|
|
nfsrv_mkdir(nfsd, mrep, md, dpos, cred, nam, mrq)
|
|
struct nfsd *nfsd;
|
|
struct mbuf *mrep, *md;
|
|
caddr_t dpos;
|
|
struct ucred *cred;
|
|
struct mbuf *nam, **mrq;
|
|
{
|
|
struct vattr va;
|
|
register struct vattr *vap = &va;
|
|
register struct nfsv2_fattr *fp;
|
|
struct nameidata nd;
|
|
register caddr_t cp;
|
|
register u_long *tl;
|
|
register long t1;
|
|
caddr_t bpos;
|
|
int error = 0, cache, len;
|
|
char *cp2;
|
|
struct mbuf *mb, *mb2, *mreq;
|
|
struct vnode *vp;
|
|
nfsv2fh_t nfh;
|
|
fhandle_t *fhp;
|
|
u_quad_t frev;
|
|
|
|
fhp = &nfh.fh_generic;
|
|
nfsm_srvmtofh(fhp);
|
|
nfsm_srvstrsiz(len, NFS_MAXNAMLEN);
|
|
nd.ni_cnd.cn_cred = cred;
|
|
nd.ni_cnd.cn_nameiop = CREATE;
|
|
nd.ni_cnd.cn_flags = LOCKPARENT;
|
|
error = nfs_namei(&nd, fhp, len, nfsd->nd_slp, nam, &md, &dpos,
|
|
nfsd->nd_procp);
|
|
if (error)
|
|
nfsm_reply(0);
|
|
nfsm_dissect(tl, u_long *, NFSX_UNSIGNED);
|
|
VATTR_NULL(vap);
|
|
vap->va_type = VDIR;
|
|
vap->va_mode = nfstov_mode(*tl++);
|
|
vp = nd.ni_vp;
|
|
if (vp != NULL) {
|
|
VOP_ABORTOP(nd.ni_dvp, &nd.ni_cnd);
|
|
if (nd.ni_dvp == vp)
|
|
vrele(nd.ni_dvp);
|
|
else
|
|
vput(nd.ni_dvp);
|
|
vrele(vp);
|
|
error = EEXIST;
|
|
nfsm_reply(0);
|
|
}
|
|
nqsrv_getl(nd.ni_dvp, NQL_WRITE);
|
|
error = VOP_MKDIR(nd.ni_dvp, &nd.ni_vp, &nd.ni_cnd, vap);
|
|
if (error)
|
|
nfsm_reply(0);
|
|
vp = nd.ni_vp;
|
|
bzero((caddr_t)fhp, sizeof(nfh));
|
|
fhp->fh_fsid = vp->v_mount->mnt_stat.f_fsid;
|
|
error = VFS_VPTOFH(vp, &fhp->fh_fid);
|
|
if (error) {
|
|
vput(vp);
|
|
nfsm_reply(0);
|
|
}
|
|
error = VOP_GETATTR(vp, vap, cred, nfsd->nd_procp);
|
|
vput(vp);
|
|
nfsm_reply(NFSX_FH+NFSX_FATTR(nfsd->nd_nqlflag != NQL_NOVAL));
|
|
nfsm_srvfhtom(fhp);
|
|
nfsm_build(fp, struct nfsv2_fattr *, NFSX_FATTR(nfsd->nd_nqlflag != NQL_NOVAL));
|
|
nfsm_srvfillattr;
|
|
return (error);
|
|
nfsmout:
|
|
VOP_ABORTOP(nd.ni_dvp, &nd.ni_cnd);
|
|
if (nd.ni_dvp == nd.ni_vp)
|
|
vrele(nd.ni_dvp);
|
|
else
|
|
vput(nd.ni_dvp);
|
|
if (nd.ni_vp)
|
|
vrele(nd.ni_vp);
|
|
return (error);
|
|
}
|
|
|
|
/*
|
|
* nfs rmdir service
|
|
*/
|
|
int
|
|
nfsrv_rmdir(nfsd, mrep, md, dpos, cred, nam, mrq)
|
|
struct nfsd *nfsd;
|
|
struct mbuf *mrep, *md;
|
|
caddr_t dpos;
|
|
struct ucred *cred;
|
|
struct mbuf *nam, **mrq;
|
|
{
|
|
register u_long *tl;
|
|
register long t1;
|
|
caddr_t bpos;
|
|
int error = 0, cache, len;
|
|
char *cp2;
|
|
struct mbuf *mb, *mreq;
|
|
struct vnode *vp;
|
|
nfsv2fh_t nfh;
|
|
fhandle_t *fhp;
|
|
struct nameidata nd;
|
|
u_quad_t frev;
|
|
|
|
fhp = &nfh.fh_generic;
|
|
nfsm_srvmtofh(fhp);
|
|
nfsm_srvstrsiz(len, NFS_MAXNAMLEN);
|
|
nd.ni_cnd.cn_cred = cred;
|
|
nd.ni_cnd.cn_nameiop = DELETE;
|
|
nd.ni_cnd.cn_flags = LOCKPARENT | LOCKLEAF;
|
|
error = nfs_namei(&nd, fhp, len, nfsd->nd_slp, nam, &md, &dpos,
|
|
nfsd->nd_procp);
|
|
if (error)
|
|
nfsm_reply(0);
|
|
vp = nd.ni_vp;
|
|
if (vp->v_type != VDIR) {
|
|
error = ENOTDIR;
|
|
goto out;
|
|
}
|
|
/*
|
|
* No rmdir "." please.
|
|
*/
|
|
if (nd.ni_dvp == vp) {
|
|
error = EINVAL;
|
|
goto out;
|
|
}
|
|
/*
|
|
* The root of a mounted filesystem cannot be deleted.
|
|
*/
|
|
if (vp->v_flag & VROOT)
|
|
error = EBUSY;
|
|
out:
|
|
if (!error) {
|
|
nqsrv_getl(nd.ni_dvp, NQL_WRITE);
|
|
nqsrv_getl(vp, NQL_WRITE);
|
|
error = VOP_RMDIR(nd.ni_dvp, nd.ni_vp, &nd.ni_cnd);
|
|
} else {
|
|
VOP_ABORTOP(nd.ni_dvp, &nd.ni_cnd);
|
|
if (nd.ni_dvp == nd.ni_vp)
|
|
vrele(nd.ni_dvp);
|
|
else
|
|
vput(nd.ni_dvp);
|
|
vput(vp);
|
|
}
|
|
nfsm_reply(0);
|
|
nfsm_srvdone;
|
|
}
|
|
|
|
/*
|
|
* nfs readdir service
|
|
* - mallocs what it thinks is enough to read
|
|
* count rounded up to a multiple of NFS_DIRBLKSIZ <= NFS_MAXREADDIR
|
|
* - calls VOP_READDIR()
|
|
* - loops around building the reply
|
|
* if the output generated exceeds count break out of loop
|
|
* The nfsm_clget macro is used here so that the reply will be packed
|
|
* tightly in mbuf clusters.
|
|
* - it only knows that it has encountered eof when the VOP_READDIR()
|
|
* reads nothing
|
|
* - as such one readdir rpc will return eof false although you are there
|
|
* and then the next will return eof
|
|
* - it trims out records with d_fileno == 0
|
|
* this doesn't matter for Unix clients, but they might confuse clients
|
|
* for other os'.
|
|
* NB: It is tempting to set eof to true if the VOP_READDIR() reads less
|
|
* than requested, but this may not apply to all filesystems. For
|
|
* example, client NFS does not { although it is never remote mounted
|
|
* anyhow }
|
|
* The alternate call nqnfsrv_readdirlook() does lookups as well.
|
|
* PS: The NFS protocol spec. does not clarify what the "count" byte
|
|
* argument is a count of.. just name strings and file id's or the
|
|
* entire reply rpc or ...
|
|
* I tried just file name and id sizes and it confused the Sun client,
|
|
* so I am using the full rpc size now. The "paranoia.." comment refers
|
|
* to including the status longwords that are not a part of the dir.
|
|
* "entry" structures, but are in the rpc.
|
|
*/
|
|
struct flrep {
|
|
u_long fl_cachable;
|
|
u_long fl_duration;
|
|
u_long fl_frev[2];
|
|
nfsv2fh_t fl_nfh;
|
|
u_long fl_fattr[NFSX_NQFATTR / sizeof (u_long)];
|
|
};
|
|
|
|
int
|
|
nfsrv_readdir(nfsd, mrep, md, dpos, cred, nam, mrq)
|
|
struct nfsd *nfsd;
|
|
struct mbuf *mrep, *md;
|
|
caddr_t dpos;
|
|
struct ucred *cred;
|
|
struct mbuf *nam, **mrq;
|
|
{
|
|
register char *bp, *be;
|
|
register struct mbuf *mp;
|
|
register struct dirent *dp;
|
|
register caddr_t cp;
|
|
register u_long *tl;
|
|
register long t1;
|
|
caddr_t bpos;
|
|
struct mbuf *mb, *mb2, *mreq, *mp2;
|
|
char *cpos, *cend, *cp2, *rbuf;
|
|
struct vnode *vp;
|
|
nfsv2fh_t nfh;
|
|
fhandle_t *fhp;
|
|
struct uio io;
|
|
struct iovec iv;
|
|
int len, nlen, rem, xfer, tsiz, i, error = 0;
|
|
int siz, cnt, fullsiz, eofflag, rdonly, cache;
|
|
u_quad_t frev;
|
|
u_long off, toff;
|
|
int ncookies;
|
|
u_int *cookies;
|
|
u_int *cookiep;
|
|
|
|
fhp = &nfh.fh_generic;
|
|
nfsm_srvmtofh(fhp);
|
|
nfsm_dissect(tl, u_long *, 2*NFSX_UNSIGNED);
|
|
off = toff = fxdr_unsigned(u_long, *tl++);
|
|
cnt = fxdr_unsigned(int, *tl);
|
|
siz = ((cnt+NFS_DIRBLKSIZ-1) & ~(NFS_DIRBLKSIZ-1));
|
|
if (cnt > NFS_MAXREADDIR)
|
|
siz = NFS_MAXREADDIR;
|
|
fullsiz = siz;
|
|
error = nfsrv_fhtovp(fhp, TRUE, &vp, cred, nfsd->nd_slp, nam, &rdonly);
|
|
if (error)
|
|
nfsm_reply(0);
|
|
nqsrv_getl(vp, NQL_READ);
|
|
error = nfsrv_access(vp, VEXEC, cred, rdonly, nfsd->nd_procp);
|
|
if (error) {
|
|
vput(vp);
|
|
nfsm_reply(0);
|
|
}
|
|
VOP_UNLOCK(vp);
|
|
MALLOC(rbuf, caddr_t, siz, M_TEMP, M_WAITOK);
|
|
again:
|
|
iv.iov_base = rbuf;
|
|
iv.iov_len = fullsiz;
|
|
io.uio_iov = &iv;
|
|
io.uio_iovcnt = 1;
|
|
io.uio_offset = (off_t)off;
|
|
io.uio_resid = fullsiz;
|
|
io.uio_segflg = UIO_SYSSPACE;
|
|
io.uio_rw = UIO_READ;
|
|
io.uio_procp = (struct proc *)0;
|
|
cookies = NULL;
|
|
eofflag = 0;
|
|
error = VOP_READDIR(vp, &io, cred, &eofflag, &ncookies, &cookies);
|
|
off = (off_t)io.uio_offset;
|
|
if (error) {
|
|
vrele(vp);
|
|
free((caddr_t)rbuf, M_TEMP);
|
|
nfsm_reply(0);
|
|
}
|
|
if (cookies == NULL) {
|
|
/*
|
|
* If the filesystem doen't support cookies, return eof.
|
|
*/
|
|
vrele(vp);
|
|
nfsm_reply(2*NFSX_UNSIGNED);
|
|
nfsm_build(tl, u_long *, 2*NFSX_UNSIGNED);
|
|
*tl++ = nfs_false;
|
|
*tl = nfs_true;
|
|
FREE((caddr_t)rbuf, M_TEMP);
|
|
return (0);
|
|
}
|
|
if (io.uio_resid) {
|
|
siz -= io.uio_resid;
|
|
|
|
/*
|
|
* If nothing read, return eof
|
|
* rpc reply
|
|
*/
|
|
if (siz == 0) {
|
|
vrele(vp);
|
|
nfsm_reply(2*NFSX_UNSIGNED);
|
|
nfsm_build(tl, u_long *, 2*NFSX_UNSIGNED);
|
|
*tl++ = nfs_false;
|
|
*tl = nfs_true;
|
|
FREE((caddr_t)rbuf, M_TEMP);
|
|
FREE(cookies, M_TEMP);
|
|
return (0);
|
|
}
|
|
}
|
|
|
|
/*
|
|
* Check for degenerate cases of nothing useful read.
|
|
* If so go try again
|
|
*/
|
|
cpos = rbuf;
|
|
cend = rbuf + siz;
|
|
dp = (struct dirent *)cpos;
|
|
cookiep = cookies;
|
|
while (cpos < cend && ncookies > 0
|
|
&& (*cookiep <= toff || dp->d_fileno == 0)) {
|
|
cpos += dp->d_reclen;
|
|
dp = (struct dirent *)cpos;
|
|
cookiep++;
|
|
ncookies--;
|
|
}
|
|
if (cpos >= cend) {
|
|
toff = off;
|
|
siz = fullsiz;
|
|
FREE(cookies, M_TEMP);
|
|
goto again;
|
|
}
|
|
|
|
len = 3*NFSX_UNSIGNED; /* paranoia, probably can be 0 */
|
|
nfsm_reply(siz);
|
|
mp = mp2 = mb;
|
|
bp = bpos;
|
|
be = bp + M_TRAILINGSPACE(mp);
|
|
|
|
/* Loop through the records and build reply */
|
|
while (cpos < cend) {
|
|
if (dp->d_fileno != 0) {
|
|
nlen = dp->d_namlen;
|
|
rem = nfsm_rndup(nlen)-nlen;
|
|
len += (4*NFSX_UNSIGNED + nlen + rem);
|
|
if (len > cnt) {
|
|
eofflag = 0;
|
|
break;
|
|
}
|
|
/*
|
|
* Build the directory record xdr from
|
|
* the dirent entry.
|
|
*/
|
|
nfsm_clget;
|
|
*tl = nfs_true;
|
|
bp += NFSX_UNSIGNED;
|
|
nfsm_clget;
|
|
*tl = txdr_unsigned(dp->d_fileno);
|
|
bp += NFSX_UNSIGNED;
|
|
nfsm_clget;
|
|
*tl = txdr_unsigned(nlen);
|
|
bp += NFSX_UNSIGNED;
|
|
|
|
/* And loop around copying the name */
|
|
xfer = nlen;
|
|
cp = dp->d_name;
|
|
while (xfer > 0) {
|
|
nfsm_clget;
|
|
if ((bp+xfer) > be)
|
|
tsiz = be-bp;
|
|
else
|
|
tsiz = xfer;
|
|
bcopy(cp, bp, tsiz);
|
|
bp += tsiz;
|
|
xfer -= tsiz;
|
|
if (xfer > 0)
|
|
cp += tsiz;
|
|
}
|
|
/* And null pad to a long boundary */
|
|
for (i = 0; i < rem; i++)
|
|
*bp++ = '\0';
|
|
nfsm_clget;
|
|
|
|
/* Finish off the record */
|
|
*tl = txdr_unsigned(*cookiep);
|
|
bp += NFSX_UNSIGNED;
|
|
}
|
|
cpos += dp->d_reclen;
|
|
dp = (struct dirent *)cpos;
|
|
cookiep++;
|
|
}
|
|
vrele(vp);
|
|
nfsm_clget;
|
|
*tl = nfs_false;
|
|
bp += NFSX_UNSIGNED;
|
|
nfsm_clget;
|
|
if (eofflag)
|
|
*tl = nfs_true;
|
|
else
|
|
*tl = nfs_false;
|
|
bp += NFSX_UNSIGNED;
|
|
if (mp != mb) {
|
|
if (bp < be)
|
|
mp->m_len = bp - mtod(mp, caddr_t);
|
|
} else
|
|
mp->m_len += bp - bpos;
|
|
FREE(rbuf, M_TEMP);
|
|
FREE(cookies, M_TEMP);
|
|
nfsm_srvdone;
|
|
}
|
|
|
|
int
|
|
nqnfsrv_readdirlook(nfsd, mrep, md, dpos, cred, nam, mrq)
|
|
struct nfsd *nfsd;
|
|
struct mbuf *mrep, *md;
|
|
caddr_t dpos;
|
|
struct ucred *cred;
|
|
struct mbuf *nam, **mrq;
|
|
{
|
|
register char *bp, *be;
|
|
register struct mbuf *mp;
|
|
register struct dirent *dp;
|
|
register caddr_t cp;
|
|
register u_long *tl;
|
|
register long t1;
|
|
caddr_t bpos;
|
|
struct mbuf *mb, *mb2, *mreq, *mp2;
|
|
char *cpos, *cend, *cp2, *rbuf;
|
|
struct vnode *vp, *nvp;
|
|
struct flrep fl;
|
|
nfsv2fh_t nfh;
|
|
fhandle_t *fhp;
|
|
struct uio io;
|
|
struct iovec iv;
|
|
struct vattr va, *vap = &va;
|
|
struct nfsv2_fattr *fp;
|
|
int len, nlen, rem, xfer, tsiz, i, error = 0, cache2;
|
|
int siz, cnt, fullsiz, eofflag, rdonly, cache;
|
|
u_long duration2;
|
|
u_quad_t frev, frev2;
|
|
u_long off, toff;
|
|
int ncookies;
|
|
u_int *cookies;
|
|
u_int *cookiep;
|
|
|
|
fhp = &nfh.fh_generic;
|
|
nfsm_srvmtofh(fhp);
|
|
nfsm_dissect(tl, u_long *, 3*NFSX_UNSIGNED);
|
|
toff = off = fxdr_unsigned(u_long, *tl++);
|
|
cnt = fxdr_unsigned(int, *tl++);
|
|
duration2 = fxdr_unsigned(int, *tl);
|
|
siz = ((cnt+NFS_DIRBLKSIZ-1) & ~(NFS_DIRBLKSIZ-1));
|
|
if (cnt > NFS_MAXREADDIR)
|
|
siz = NFS_MAXREADDIR;
|
|
fullsiz = siz;
|
|
error = nfsrv_fhtovp(fhp, TRUE, &vp, cred, nfsd->nd_slp, nam, &rdonly);
|
|
if (error)
|
|
nfsm_reply(0);
|
|
nqsrv_getl(vp, NQL_READ);
|
|
error = nfsrv_access(vp, VEXEC, cred, rdonly, nfsd->nd_procp);
|
|
if (error) {
|
|
vput(vp);
|
|
nfsm_reply(0);
|
|
}
|
|
VOP_UNLOCK(vp);
|
|
MALLOC(rbuf, caddr_t, siz, M_TEMP, M_WAITOK);
|
|
again:
|
|
iv.iov_base = rbuf;
|
|
iv.iov_len = fullsiz;
|
|
io.uio_iov = &iv;
|
|
io.uio_iovcnt = 1;
|
|
io.uio_offset = (off_t)off;
|
|
io.uio_resid = fullsiz;
|
|
io.uio_segflg = UIO_SYSSPACE;
|
|
io.uio_rw = UIO_READ;
|
|
io.uio_procp = (struct proc *)0;
|
|
cookies = NULL;
|
|
eofflag = 0;
|
|
error = VOP_READDIR(vp, &io, cred, &eofflag, &ncookies, &cookies);
|
|
off = (u_long)io.uio_offset;
|
|
if (error) {
|
|
vrele(vp);
|
|
free((caddr_t)rbuf, M_TEMP);
|
|
nfsm_reply(0);
|
|
}
|
|
if (cookies == NULL) {
|
|
/*
|
|
* If the filesystem doen't support cookies, return eof.
|
|
*/
|
|
vrele(vp);
|
|
nfsm_reply(2*NFSX_UNSIGNED);
|
|
nfsm_build(tl, u_long *, 2*NFSX_UNSIGNED);
|
|
*tl++ = nfs_false;
|
|
*tl = nfs_true;
|
|
FREE((caddr_t)rbuf, M_TEMP);
|
|
return (0);
|
|
}
|
|
if (io.uio_resid) {
|
|
siz -= io.uio_resid;
|
|
|
|
/*
|
|
* If nothing read, return eof
|
|
* rpc reply
|
|
*/
|
|
if (siz == 0) {
|
|
vrele(vp);
|
|
nfsm_reply(2 * NFSX_UNSIGNED);
|
|
nfsm_build(tl, u_long *, 2 * NFSX_UNSIGNED);
|
|
*tl++ = nfs_false;
|
|
*tl = nfs_true;
|
|
FREE((caddr_t)rbuf, M_TEMP);
|
|
FREE(cookies, M_TEMP);
|
|
return (0);
|
|
}
|
|
}
|
|
|
|
/*
|
|
* Check for degenerate cases of nothing useful read.
|
|
* If so go try again
|
|
*/
|
|
cpos = rbuf;
|
|
cend = rbuf + siz;
|
|
dp = (struct dirent *)cpos;
|
|
cookiep = cookies;
|
|
while (cpos < cend && ncookies > 0
|
|
&& (*cookiep <= toff || dp->d_fileno == 0)) {
|
|
cpos += dp->d_reclen;
|
|
dp = (struct dirent *)cpos;
|
|
cookiep++;
|
|
ncookies--;
|
|
}
|
|
if (cpos >= cend) {
|
|
toff = off;
|
|
siz = fullsiz;
|
|
FREE(cookies, M_TEMP);
|
|
goto again;
|
|
}
|
|
|
|
len = 3 * NFSX_UNSIGNED; /* paranoia, probably can be 0 */
|
|
nfsm_reply(siz);
|
|
mp = mp2 = mb;
|
|
bp = bpos;
|
|
be = bp + M_TRAILINGSPACE(mp);
|
|
|
|
/* Loop through the records and build reply */
|
|
while (cpos < cend) {
|
|
if (dp->d_fileno != 0) {
|
|
nlen = dp->d_namlen;
|
|
rem = nfsm_rndup(nlen)-nlen;
|
|
|
|
/*
|
|
* For readdir_and_lookup get the vnode using
|
|
* the file number.
|
|
*/
|
|
if (VFS_VGET(vp->v_mount, dp->d_fileno, &nvp))
|
|
goto invalid;
|
|
bzero((caddr_t)&fl.fl_nfh, sizeof (nfsv2fh_t));
|
|
fl.fl_nfh.fh_generic.fh_fsid =
|
|
nvp->v_mount->mnt_stat.f_fsid;
|
|
if (VFS_VPTOFH(nvp, &fl.fl_nfh.fh_generic.fh_fid)) {
|
|
vput(nvp);
|
|
goto invalid;
|
|
}
|
|
if (duration2) {
|
|
(void) nqsrv_getlease(nvp, &duration2, NQL_READ,
|
|
nfsd, nam, &cache2, &frev2, cred);
|
|
fl.fl_duration = txdr_unsigned(duration2);
|
|
fl.fl_cachable = txdr_unsigned(cache2);
|
|
txdr_hyper(&frev2, fl.fl_frev);
|
|
} else
|
|
fl.fl_duration = 0;
|
|
if (VOP_GETATTR(nvp, vap, cred, nfsd->nd_procp)) {
|
|
vput(nvp);
|
|
goto invalid;
|
|
}
|
|
vput(nvp);
|
|
fp = (struct nfsv2_fattr *)&fl.fl_fattr;
|
|
nfsm_srvfillattr;
|
|
len += (4*NFSX_UNSIGNED + nlen + rem + NFSX_FH
|
|
+ NFSX_NQFATTR);
|
|
if (len > cnt) {
|
|
eofflag = 0;
|
|
break;
|
|
}
|
|
/*
|
|
* Build the directory record xdr from
|
|
* the dirent entry.
|
|
*/
|
|
nfsm_clget;
|
|
*tl = nfs_true;
|
|
bp += NFSX_UNSIGNED;
|
|
|
|
/*
|
|
* For readdir_and_lookup copy the stuff out.
|
|
*/
|
|
xfer = sizeof (struct flrep);
|
|
cp = (caddr_t)&fl;
|
|
while (xfer > 0) {
|
|
nfsm_clget;
|
|
if ((bp+xfer) > be)
|
|
tsiz = be-bp;
|
|
else
|
|
tsiz = xfer;
|
|
bcopy(cp, bp, tsiz);
|
|
bp += tsiz;
|
|
xfer -= tsiz;
|
|
if (xfer > 0)
|
|
cp += tsiz;
|
|
}
|
|
nfsm_clget;
|
|
*tl = txdr_unsigned(dp->d_fileno);
|
|
bp += NFSX_UNSIGNED;
|
|
nfsm_clget;
|
|
*tl = txdr_unsigned(nlen);
|
|
bp += NFSX_UNSIGNED;
|
|
|
|
/* And loop around copying the name */
|
|
xfer = nlen;
|
|
cp = dp->d_name;
|
|
while (xfer > 0) {
|
|
nfsm_clget;
|
|
if ((bp+xfer) > be)
|
|
tsiz = be-bp;
|
|
else
|
|
tsiz = xfer;
|
|
bcopy(cp, bp, tsiz);
|
|
bp += tsiz;
|
|
xfer -= tsiz;
|
|
if (xfer > 0)
|
|
cp += tsiz;
|
|
}
|
|
/* And null pad to a long boundary */
|
|
for (i = 0; i < rem; i++)
|
|
*bp++ = '\0';
|
|
nfsm_clget;
|
|
|
|
/* Finish off the record */
|
|
*tl = txdr_unsigned(*cookiep);
|
|
bp += NFSX_UNSIGNED;
|
|
}
|
|
invalid:
|
|
cpos += dp->d_reclen;
|
|
dp = (struct dirent *)cpos;
|
|
cookiep++;
|
|
}
|
|
vrele(vp);
|
|
nfsm_clget;
|
|
*tl = nfs_false;
|
|
bp += NFSX_UNSIGNED;
|
|
nfsm_clget;
|
|
if (eofflag)
|
|
*tl = nfs_true;
|
|
else
|
|
*tl = nfs_false;
|
|
bp += NFSX_UNSIGNED;
|
|
if (mp != mb) {
|
|
if (bp < be)
|
|
mp->m_len = bp - mtod(mp, caddr_t);
|
|
} else
|
|
mp->m_len += bp - bpos;
|
|
FREE(rbuf, M_TEMP);
|
|
FREE(cookies, M_TEMP);
|
|
nfsm_srvdone;
|
|
}
|
|
|
|
/*
|
|
* nfs statfs service
|
|
*/
|
|
int
|
|
nfsrv_statfs(nfsd, mrep, md, dpos, cred, nam, mrq)
|
|
struct nfsd *nfsd;
|
|
struct mbuf *mrep, *md;
|
|
caddr_t dpos;
|
|
struct ucred *cred;
|
|
struct mbuf *nam, **mrq;
|
|
{
|
|
register struct statfs *sf;
|
|
register struct nfsv2_statfs *sfp;
|
|
register u_long *tl;
|
|
register long t1;
|
|
caddr_t bpos;
|
|
int error = 0, rdonly, cache = 0, isnq;
|
|
char *cp2;
|
|
struct mbuf *mb, *mb2, *mreq;
|
|
struct vnode *vp;
|
|
nfsv2fh_t nfh;
|
|
fhandle_t *fhp;
|
|
struct statfs statfs;
|
|
u_quad_t frev;
|
|
|
|
fhp = &nfh.fh_generic;
|
|
isnq = (nfsd->nd_nqlflag != NQL_NOVAL);
|
|
nfsm_srvmtofh(fhp);
|
|
error = nfsrv_fhtovp(fhp, TRUE, &vp, cred, nfsd->nd_slp, nam, &rdonly);
|
|
if (error)
|
|
nfsm_reply(0);
|
|
sf = &statfs;
|
|
error = VFS_STATFS(vp->v_mount, sf, nfsd->nd_procp);
|
|
vput(vp);
|
|
nfsm_reply(NFSX_STATFS(isnq));
|
|
nfsm_build(sfp, struct nfsv2_statfs *, NFSX_STATFS(isnq));
|
|
sfp->sf_tsize = txdr_unsigned(NFS_MAXDGRAMDATA);
|
|
sfp->sf_bsize = txdr_unsigned(sf->f_bsize);
|
|
sfp->sf_blocks = txdr_unsigned(sf->f_blocks);
|
|
sfp->sf_bfree = txdr_unsigned(sf->f_bfree);
|
|
sfp->sf_bavail = txdr_unsigned(sf->f_bavail);
|
|
if (isnq) {
|
|
sfp->sf_files = txdr_unsigned(sf->f_files);
|
|
sfp->sf_ffree = txdr_unsigned(sf->f_ffree);
|
|
}
|
|
nfsm_srvdone;
|
|
}
|
|
|
|
/*
|
|
* Null operation, used by clients to ping server
|
|
*/
|
|
/* ARGSUSED */
|
|
int
|
|
nfsrv_null(nfsd, mrep, md, dpos, cred, nam, mrq)
|
|
struct nfsd *nfsd;
|
|
struct mbuf *mrep, *md;
|
|
caddr_t dpos;
|
|
struct ucred *cred;
|
|
struct mbuf *nam, **mrq;
|
|
{
|
|
caddr_t bpos;
|
|
int error = VNOVAL, cache = 0;
|
|
struct mbuf *mb, *mreq;
|
|
u_quad_t frev;
|
|
|
|
nfsm_reply(0);
|
|
return (error);
|
|
}
|
|
|
|
/*
|
|
* No operation, used for obsolete procedures
|
|
*/
|
|
/* ARGSUSED */
|
|
int
|
|
nfsrv_noop(nfsd, mrep, md, dpos, cred, nam, mrq)
|
|
struct nfsd *nfsd;
|
|
struct mbuf *mrep, *md;
|
|
caddr_t dpos;
|
|
struct ucred *cred;
|
|
struct mbuf *nam, **mrq;
|
|
{
|
|
caddr_t bpos;
|
|
int error, cache = 0;
|
|
struct mbuf *mb, *mreq;
|
|
u_quad_t frev;
|
|
|
|
if (nfsd->nd_repstat)
|
|
error = nfsd->nd_repstat;
|
|
else
|
|
error = EPROCUNAVAIL;
|
|
nfsm_reply(0);
|
|
return (error);
|
|
}
|
|
|
|
/*
|
|
* Perform access checking for vnodes obtained from file handles that would
|
|
* refer to files already opened by a Unix client. You cannot just use
|
|
* vn_writechk() and VOP_ACCESS() for two reasons.
|
|
* 1 - You must check for exported rdonly as well as MNT_RDONLY for the write case
|
|
* 2 - The owner is to be given access irrespective of mode bits so that
|
|
* processes that chmod after opening a file don't break. I don't like
|
|
* this because it opens a security hole, but since the nfs server opens
|
|
* a security hole the size of a barn door anyhow, what the heck.
|
|
*/
|
|
int
|
|
nfsrv_access(vp, flags, cred, rdonly, p)
|
|
register struct vnode *vp;
|
|
int flags;
|
|
register struct ucred *cred;
|
|
int rdonly;
|
|
struct proc *p;
|
|
{
|
|
struct vattr vattr;
|
|
int error;
|
|
if (flags & VWRITE) {
|
|
/* Just vn_writechk() changed to check rdonly */
|
|
/*
|
|
* Disallow write attempts on read-only file systems;
|
|
* unless the file is a socket or a block or character
|
|
* device resident on the file system.
|
|
*/
|
|
if (rdonly || (vp->v_mount->mnt_flag & MNT_RDONLY)) {
|
|
switch (vp->v_type) {
|
|
case VREG: case VDIR: case VLNK:
|
|
return (EROFS);
|
|
default:
|
|
break;
|
|
}
|
|
}
|
|
/*
|
|
* If there's shared text associated with
|
|
* the inode, try to free it up once. If
|
|
* we fail, we can't allow writing.
|
|
*/
|
|
if ((vp->v_flag & VTEXT) && !vnode_pager_uncache(vp))
|
|
return (ETXTBSY);
|
|
}
|
|
error = VOP_GETATTR(vp, &vattr, cred, p);
|
|
if (error)
|
|
return (error);
|
|
if ((error = VOP_ACCESS(vp, flags, cred, p)) &&
|
|
cred->cr_uid != vattr.va_uid)
|
|
return (error);
|
|
return (0);
|
|
}
|