]> git.saurik.com Git - apple/xnu.git/blobdiff - bsd/vfs/vfs_vnops.c
xnu-2050.7.9.tar.gz
[apple/xnu.git] / bsd / vfs / vfs_vnops.c
index e186fb48408a751e122d7e59d833bc97b3bd8e89..47552a0d8e0a33eb23083bcee7560767b6e26a5b 100644 (file)
@@ -1,23 +1,29 @@
 /*
- * Copyright (c) 2000-2002 Apple Computer, Inc. All rights reserved.
+ * Copyright (c) 2000-2012 Apple Inc. All rights reserved.
  *
- * @APPLE_LICENSE_HEADER_START@
+ * @APPLE_OSREFERENCE_LICENSE_HEADER_START@
  * 
- * The contents of this file constitute Original Code as defined in and
- * are subject to the Apple Public Source License Version 1.1 (the
- * "License").  You may not use this file except in compliance with the
- * License.  Please obtain a copy of the License at
- * http://www.apple.com/publicsource and read it before using this file.
+ * This file contains Original Code and/or Modifications of Original Code
+ * as defined in and that are subject to the Apple Public Source License
+ * Version 2.0 (the 'License'). You may not use this file except in
+ * compliance with the License. The rights granted to you under the License
+ * may not be used to create, or enable the creation or redistribution of,
+ * unlawful or unlicensed copies of an Apple operating system, or to
+ * circumvent, violate, or enable the circumvention or violation of, any
+ * terms of an Apple operating system software license agreement.
  * 
- * This Original Code and all software distributed under the License are
- * distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY KIND, EITHER
+ * Please obtain a copy of the License at
+ * http://www.opensource.apple.com/apsl/ and read it before using this file.
+ * 
+ * The Original Code and all software distributed under the License are
+ * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
  * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
  * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE OR NON-INFRINGEMENT.  Please see the
- * License for the specific language governing rights and limitations
- * under the License.
+ * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
+ * Please see the License for the specific language governing rights and
+ * limitations under the License.
  * 
- * @APPLE_LICENSE_HEADER_END@
+ * @APPLE_OSREFERENCE_LICENSE_HEADER_END@
  */
 /* Copyright (c) 1995 NeXT Computer, Inc. All Rights Reserved */
 /*
  *     @(#)vfs_vnops.c 8.14 (Berkeley) 6/15/95
  *
  */
+/*
+ * NOTICE: This file was modified by SPARTA, Inc. in 2005 to introduce
+ * support for mandatory and extensible security protections.  This notice
+ * is included in support of clause 2.2 (b) of the Apple Public License,
+ * Version 2.0.
+ */
 
 #include <sys/param.h>
+#include <sys/types.h>
 #include <sys/systm.h>
 #include <sys/kernel.h>
-#include <sys/file.h>
+#include <sys/file_internal.h>
 #include <sys/stat.h>
-#include <sys/buf.h>
-#include <sys/proc.h>
-#include <sys/mount.h>
+#include <sys/proc_internal.h>
+#include <sys/kauth.h>
+#include <sys/mount_internal.h>
 #include <sys/namei.h>
-#include <sys/vnode.h>
+#include <sys/vnode_internal.h>
 #include <sys/ioctl.h>
 #include <sys/tty.h>
+/* Temporary workaround for ubc.h until <rdar://4714366 is resolved */
+#define ubc_setcred ubc_setcred_deprecated
 #include <sys/ubc.h>
+#undef ubc_setcred
+int    ubc_setcred(struct vnode *, struct proc *);
 #include <sys/conf.h>
 #include <sys/disk.h>
+#include <sys/fsevents.h>
+#include <sys/kdebug.h>
+#include <sys/xattr.h>
+#include <sys/ubc_internal.h>
+#include <sys/uio_internal.h>
+#include <sys/resourcevar.h>
+#include <sys/signalvar.h>
 
 #include <vm/vm_kern.h>
+#include <vm/vm_map.h>
 
 #include <miscfs/specfs/specdev.h>
+#include <miscfs/fifofs/fifo.h>
+
+#if CONFIG_MACF
+#include <security/mac_framework.h>
+#endif
 
-static int vn_closefile __P((struct file *fp, struct proc *p));
-static int vn_ioctl __P((struct file *fp, u_long com, caddr_t data,
-               struct proc *p));
-static int vn_read __P((struct file *fp, struct uio *uio,
-               struct ucred *cred, int flags, struct proc *p));
-static int vn_write __P((struct file *fp, struct uio *uio,
-               struct ucred *cred, int flags, struct proc *p));
-static int vn_select __P(( struct file *fp, int which, void * wql,
-               struct proc *p));
+#if CONFIG_PROTECT
+#include <sys/cprotect.h>
+#endif
+
+
+static int vn_closefile(struct fileglob *fp, vfs_context_t ctx);
+static int vn_ioctl(struct fileproc *fp, u_long com, caddr_t data,
+                       vfs_context_t ctx);
+static int vn_read(struct fileproc *fp, struct uio *uio, int flags,
+                       vfs_context_t ctx);
+static int vn_write(struct fileproc *fp, struct uio *uio, int flags,
+                       vfs_context_t ctx);
+static int vn_select( struct fileproc *fp, int which, void * wql,
+                       vfs_context_t ctx);
+static int vn_kqfilt_add(struct fileproc *fp, struct knote *kn,
+                       vfs_context_t ctx);
+static void filt_vndetach(struct knote *kn);
+static int filt_vnode(struct knote *kn, long hint);
+static int vn_open_auth_finish(vnode_t vp, int fmode, vfs_context_t ctx);
+#if 0
+static int vn_kqfilt_remove(struct vnode *vp, uintptr_t ident,
+                       vfs_context_t ctx);
+#endif
 
 struct         fileops vnops =
-       { vn_read, vn_write, vn_ioctl, vn_select, vn_closefile };
+       { vn_read, vn_write, vn_ioctl, vn_select, vn_closefile, vn_kqfilt_add, NULL };
+
+struct  filterops vnode_filtops = { 
+       .f_isfd = 1, 
+       .f_attach = NULL, 
+       .f_detach = filt_vndetach, 
+       .f_event = filt_vnode
+};
 
 /*
  * Common code for vnode open operations.
- * Check permissions, and call the VOP_OPEN or VOP_CREATE routine.
+ * Check permissions, and call the VNOP_OPEN or VNOP_CREATE routine.
+ *
+ * XXX the profusion of interfaces here is probably a bad thing.
+ */
+int
+vn_open(struct nameidata *ndp, int fmode, int cmode)
+{
+       return(vn_open_modflags(ndp, &fmode, cmode));
+}
+
+int
+vn_open_modflags(struct nameidata *ndp, int *fmodep, int cmode)
+{
+       struct vnode_attr va;
+
+       VATTR_INIT(&va);
+       VATTR_SET(&va, va_mode, cmode);
+       
+       return(vn_open_auth(ndp, fmodep, &va));
+}
+
+static int
+vn_open_auth_finish(vnode_t vp, int fmode, vfs_context_t ctx)
+{
+       int error;
+
+       if ((error = vnode_ref_ext(vp, fmode, 0)) != 0) {
+               goto bad;
+       }
+
+       /* call out to allow 3rd party notification of open. 
+        * Ignore result of kauth_authorize_fileop call.
+        */
+       kauth_authorize_fileop(vfs_context_ucred(ctx), KAUTH_FILEOP_OPEN, 
+                                                  (uintptr_t)vp, 0);
+
+       return 0;
+
+bad:
+       return error;
+
+}
+
+/*
+ * May do nameidone() to allow safely adding an FSEvent.  Cue off of ni_dvp to
+ * determine whether that has happened.  
+ */
+static int
+vn_open_auth_do_create(struct nameidata *ndp, struct vnode_attr *vap, int fmode, boolean_t *did_create, boolean_t *did_open, vfs_context_t ctx)
+{
+       uint32_t status = 0;
+       vnode_t dvp = ndp->ni_dvp;
+       int batched;
+       int error;
+       vnode_t vp;
+
+       batched = vnode_compound_open_available(ndp->ni_dvp);
+       *did_open = FALSE;
+
+       VATTR_SET(vap, va_type, VREG);
+       if (fmode & O_EXCL)
+               vap->va_vaflags |= VA_EXCLUSIVE;
+
+#if NAMEDRSRCFORK
+       if (ndp->ni_cnd.cn_flags & CN_WANTSRSRCFORK) {
+               if ((error = vn_authorize_create(dvp, &ndp->ni_cnd, vap, ctx, NULL)) != 0) 
+                       goto out;
+               if ((error = vnode_makenamedstream(dvp, &ndp->ni_vp, XATTR_RESOURCEFORK_NAME, 0, ctx)) != 0)
+                       goto out;
+               *did_create = TRUE;
+       } else {
+#endif
+               if (!batched) {
+                       if ((error = vn_authorize_create(dvp, &ndp->ni_cnd, vap, ctx, NULL)) != 0)
+                               goto out;
+               }
+
+               error = vn_create(dvp, &ndp->ni_vp, ndp, vap, VN_CREATE_DOOPEN, fmode, &status, ctx);
+               if (error != 0) {
+                       if (batched) {
+                               *did_create = (status & COMPOUND_OPEN_STATUS_DID_CREATE) ? TRUE : FALSE;
+                       } else {
+                               *did_create = FALSE;
+                       }
+
+                       if (error == EKEEPLOOKING) {
+                               if (*did_create) {
+                                       panic("EKEEPLOOKING, but we did a create?");
+                               }
+                               if (!batched) {
+                                       panic("EKEEPLOOKING from filesystem that doesn't support compound vnops?");
+                               }
+                               if ((ndp->ni_flag & NAMEI_CONTLOOKUP) == 0) {
+                                       panic("EKEEPLOOKING, but continue flag not set?");
+                               }
+
+                               /* 
+                                * Do NOT drop the dvp: we need everything to continue the lookup.
+                                */
+                               return error;
+                       }
+               } else {
+                       if (batched) {
+                               *did_create = (status & COMPOUND_OPEN_STATUS_DID_CREATE) ? 1 : 0;
+                               *did_open = TRUE;
+                       } else {
+                               *did_create = TRUE;
+                       }
+               }
+#if NAMEDRSRCFORK
+       }
+#endif
+
+       /* 
+       * Unlock the fsnode (if locked) here so that we are free
+       * to drop the dvp iocount and prevent deadlock in build_path().
+       * nameidone() will still do the right thing later.
+       */
+       vp = ndp->ni_vp;
+       namei_unlock_fsnode(ndp);
+
+       if (*did_create) {
+               int     update_flags = 0;
+
+               // Make sure the name & parent pointers are hooked up
+               if (vp->v_name == NULL)
+                       update_flags |= VNODE_UPDATE_NAME;
+               if (vp->v_parent == NULLVP)
+                       update_flags |= VNODE_UPDATE_PARENT;
+
+               if (update_flags)
+                       vnode_update_identity(vp, dvp, ndp->ni_cnd.cn_nameptr, ndp->ni_cnd.cn_namelen, ndp->ni_cnd.cn_hash, update_flags);
+
+               vnode_put(dvp);
+               ndp->ni_dvp = NULLVP;
+
+#if CONFIG_FSE
+               if (need_fsevent(FSE_CREATE_FILE, vp)) {
+                       add_fsevent(FSE_CREATE_FILE, ctx,
+                                       FSE_ARG_VNODE, vp,
+                                       FSE_ARG_DONE);
+               }
+#endif
+       }
+out:
+       if (ndp->ni_dvp != NULLVP) {
+               vnode_put(dvp);
+               ndp->ni_dvp = NULLVP;
+       }
+
+       return error;
+}
+
+/*
+ * Open a file with authorization, updating the contents of the structures
+ * pointed to by ndp, fmodep, and vap as necessary to perform the requested
+ * operation.  This function is used for both opens of existing files, and
+ * creation of new files.
+ *
+ * Parameters: ndp                     The nami data pointer describing the
+ *                                     file
+ *             fmodep                  A pointer to an int containg the mode
+ *                                     information to be used for the open
+ *             vap                     A pointer to the vnode attribute
+ *                                     descriptor to be used for the open
+ *
+ * Indirect:   *                       Contents of the data structures pointed
+ *                                     to by the parameters are modified as
+ *                                     necessary to the requested operation.
+ *
+ * Returns:    0                       Success
+ *             !0                      errno value
+ *
+ * Notes:      The kauth_filesec_t in 'vap', if any, is in host byte order.
+ *
+ *             The contents of '*ndp' will be modified, based on the other
+ *             arguments to this function, and to return file and directory
+ *             data necessary to satisfy the requested operation.
+ *
+ *             If the file does not exist and we are creating it, then the
+ *             O_TRUNC flag will be cleared in '*fmodep' to indicate to the
+ *             caller that the file was not truncated.
+ *
+ *             If the file exists and the O_EXCL flag was not specified, then
+ *             the O_CREAT flag will be cleared in '*fmodep' to indicate to
+ *             the caller that the existing file was merely opened rather
+ *             than created.
+ *
+ *             The contents of '*vap' will be modified as necessary to
+ *             complete the operation, including setting of supported
+ *             attribute, clearing of fields containing unsupported attributes
+ *             in the request, if the request proceeds without them, etc..
+ *
+ * XXX:                This function is too complicated in actings on its arguments
+ *
+ * XXX:                We should enummerate the possible errno values here, and where
+ *             in the code they originated.
  */
 int
-vn_open(ndp, fmode, cmode)
-       register struct nameidata *ndp;
-       int fmode, cmode;
+vn_open_auth(struct nameidata *ndp, int *fmodep, struct vnode_attr *vap)
 {
-       register struct vnode *vp;
-       register struct proc *p = ndp->ni_cnd.cn_proc;
-       register struct ucred *cred = p->p_ucred;
-       struct vattr vat;
-       struct vattr *vap = &vat;
+       struct vnode *vp;
+       struct vnode *dvp;
+       vfs_context_t ctx = ndp->ni_cnd.cn_context;
        int error;
-       int didhold = 0;
+       int fmode;
+       uint32_t origcnflags;
+       boolean_t did_create;
+       boolean_t did_open;
+       boolean_t need_vnop_open;
+       boolean_t batched;
+       boolean_t ref_failed;
+
+again:
+       vp = NULL;
+       dvp = NULL;
+       batched = FALSE;
+       did_create = FALSE;
+       need_vnop_open = TRUE;
+       ref_failed = FALSE;
+       fmode = *fmodep;
+       origcnflags = ndp->ni_cnd.cn_flags;
 
+       /*
+        * O_CREAT
+        */
        if (fmode & O_CREAT) {
+               if ( (fmode & O_DIRECTORY) ) {
+                       error = EINVAL;
+                       goto out;
+               }
                ndp->ni_cnd.cn_nameiop = CREATE;
-               ndp->ni_cnd.cn_flags = LOCKPARENT | LOCKLEAF;
-               if ((fmode & O_EXCL) == 0)
+#if CONFIG_TRIGGERS
+               ndp->ni_op = OP_LINK;
+#endif
+               /* Inherit USEDVP, vnode_open() supported flags only */
+               ndp->ni_cnd.cn_flags &= (USEDVP | NOCROSSMOUNT | DOWHITEOUT);
+               ndp->ni_cnd.cn_flags |= LOCKPARENT | LOCKLEAF | AUDITVNPATH1;
+               ndp->ni_flag = NAMEI_COMPOUNDOPEN;
+#if NAMEDRSRCFORK
+               /* open calls are allowed for resource forks. */
+               ndp->ni_cnd.cn_flags |= CN_ALLOWRSRCFORK;
+#endif
+               if ((fmode & O_EXCL) == 0 && (fmode & O_NOFOLLOW) == 0 && (origcnflags & FOLLOW) != 0)
                        ndp->ni_cnd.cn_flags |= FOLLOW;
-               bwillwrite();
-               if (error = namei(ndp))
-                       return (error);
-               if (ndp->ni_vp == NULL) {
-                       VATTR_NULL(vap);
-                       vap->va_type = VREG;
-                       vap->va_mode = cmode;
-                       if (fmode & O_EXCL)
-                               vap->va_vaflags |= VA_EXCLUSIVE;
-                       VOP_LEASE(ndp->ni_dvp, p, cred, LEASE_WRITE);
-                       if (error = VOP_CREATE(ndp->ni_dvp, &ndp->ni_vp,
-                           &ndp->ni_cnd, vap))
-                               return (error);
-                       fmode &= ~O_TRUNC;
-                       vp = ndp->ni_vp;
-               } else {
-                       VOP_ABORTOP(ndp->ni_dvp, &ndp->ni_cnd);
-                       if (ndp->ni_dvp == ndp->ni_vp)
-                               vrele(ndp->ni_dvp);
-                       else
-                               vput(ndp->ni_dvp);
-                       ndp->ni_dvp = NULL;
+
+continue_create_lookup:
+               if ( (error = namei(ndp)) )
+                       goto out;
+
+               dvp = ndp->ni_dvp;
+               vp = ndp->ni_vp;
+
+               batched = vnode_compound_open_available(dvp);
+
+               /* not found, create */
+               if (vp == NULL) {
+                       /* must have attributes for a new file */
+                       if (vap == NULL) {
+                               error = EINVAL;
+                               goto out;
+                       }
+                       /*
+                        * Attempt a create.   For a system supporting compound VNOPs, we may
+                        * find an existing file or create one; in either case, we will already
+                        * have the file open and no VNOP_OPEN() will be needed.
+                        */
+                       error = vn_open_auth_do_create(ndp, vap, fmode, &did_create, &did_open, ctx);
+
+                       dvp = ndp->ni_dvp;
                        vp = ndp->ni_vp;
-                       if (fmode & O_EXCL) {
+
+                       /* 
+                        * Detected a node that the filesystem couldn't handle.  Don't call
+                        * nameidone() yet, because we need that path buffer.
+                        */
+                       if (error == EKEEPLOOKING) {
+                               if (!batched) {
+                                       panic("EKEEPLOOKING from a filesystem that doesn't support compound VNOPs?");
+                               }
+                               goto continue_create_lookup;
+                       }
+
+                       nameidone(ndp);
+                       if (dvp) {
+                               panic("Shouldn't have a dvp here.");
+                       }
+
+                       if (error) {
+                               /*
+                                * Check for a creation or unlink race.
+                                */
+                               if (((error == EEXIST) && !(fmode & O_EXCL)) ||
+                                               ((error == ENOENT) && (fmode & O_CREAT))){
+                                       if (vp) 
+                                               vnode_put(vp);
+                                       goto again;
+                               }
+                               goto bad;
+                       }
+
+                       need_vnop_open = !did_open;
+               } 
+               else {
+                       if (fmode & O_EXCL)
                                error = EEXIST;
+
+                       /* 
+                        * We have a vnode.  Use compound open if available 
+                        * or else fall through to "traditional" path.  Note: can't
+                        * do a compound open for root, because the parent belongs
+                        * to a different FS.
+                        */
+                       if (error == 0 && batched && (vnode_mount(dvp) == vnode_mount(vp))) {
+                               error = VNOP_COMPOUND_OPEN(dvp, &ndp->ni_vp, ndp, 0, fmode, NULL, NULL, ctx);
+
+                               if (error == 0) {
+                                       vp = ndp->ni_vp;
+                                       need_vnop_open = FALSE;
+                               } else if (error == EKEEPLOOKING) {
+                                       if ((ndp->ni_flag & NAMEI_CONTLOOKUP) == 0) {
+                                               panic("EKEEPLOOKING, but continue flag not set?");
+                                       }
+                                       goto continue_create_lookup;
+                               } 
+                       }
+                       nameidone(ndp);
+                       vnode_put(dvp);
+                       ndp->ni_dvp = NULLVP;
+
+                       if (error) {
                                goto bad;
                        }
+
                        fmode &= ~O_CREAT;
+
+                       /* Fall through */
                }
        } else {
+               /*
+                * Not O_CREAT
+                */
                ndp->ni_cnd.cn_nameiop = LOOKUP;
-               ndp->ni_cnd.cn_flags = FOLLOW | LOCKLEAF;
-               if (error = namei(ndp))
-                       return (error);
-               vp = ndp->ni_vp;
-       }
-       if (vp->v_type == VSOCK) {
-               error = EOPNOTSUPP;
-               goto bad;
-       }
+               /* Inherit USEDVP, vnode_open() supported flags only */
+               ndp->ni_cnd.cn_flags &= (USEDVP | NOCROSSMOUNT | DOWHITEOUT);
+               ndp->ni_cnd.cn_flags |= FOLLOW | LOCKLEAF | AUDITVNPATH1 | WANTPARENT;
+#if NAMEDRSRCFORK
+               /* open calls are allowed for resource forks. */
+               ndp->ni_cnd.cn_flags |= CN_ALLOWRSRCFORK;
+#endif
+               ndp->ni_flag = NAMEI_COMPOUNDOPEN;
 
-#if DIAGNOSTIC
-       if (UBCINFOMISSING(vp))
-               panic("vn_open: ubc_info_init");
-#endif /* DIAGNOSTIC */
+               /* preserve NOFOLLOW from vnode_open() */
+               if (fmode & O_NOFOLLOW || fmode & O_SYMLINK || (origcnflags & FOLLOW) == 0) {
+                       ndp->ni_cnd.cn_flags &= ~FOLLOW;
+               }
 
-       if (UBCINFOEXISTS(vp) && ((didhold = ubc_hold(vp)) == 0)) {
-               error = ENOENT;
-               goto bad;
+               /* Do a lookup, possibly going directly to filesystem for compound operation */
+               do {
+                       if ( (error = namei(ndp)) )
+                               goto out;
+                       vp = ndp->ni_vp;
+                       dvp = ndp->ni_dvp;
+
+                       /* Check for batched lookup-open */
+                       batched = vnode_compound_open_available(dvp);
+                       if (batched && ((vp == NULLVP) || (vnode_mount(dvp) == vnode_mount(vp)))) {
+                               error = VNOP_COMPOUND_OPEN(dvp, &ndp->ni_vp, ndp, 0, fmode, NULL, NULL, ctx);
+                               vp = ndp->ni_vp;
+                               if (error == 0) {
+                                       need_vnop_open = FALSE;
+                               } else if (error == EKEEPLOOKING) {
+                                       if ((ndp->ni_flag & NAMEI_CONTLOOKUP) == 0) {
+                                               panic("EKEEPLOOKING, but continue flag not set?");
+                                       }
+                               }
+                       }
+               } while (error == EKEEPLOOKING);
+
+               nameidone(ndp);
+               vnode_put(dvp);
+               ndp->ni_dvp = NULLVP;
+
+               if (error) {
+                       goto bad;
+               }
        }
 
-       if ((fmode & O_CREAT) == 0) {
-               if (fmode & FREAD && fmode & (FWRITE | O_TRUNC)) {
-                       int err = 0;
-                       if (vp->v_type == VDIR)
-                               err = EISDIR;
-                       else
-                               err = vn_writechk(vp);
-                       if (err && !(error = VOP_ACCESS(vp, VREAD, cred, p)))
-                               error = err;
-                       if (error || (error = VOP_ACCESS(vp, VREAD|VWRITE,
-                                                        cred, p)))
-                               goto bad;
-               } else if (fmode & FREAD) {
-                       if ((error = VOP_ACCESS(vp, VREAD, cred, p)))
-                               goto bad;
-               } else if (fmode & (FWRITE | O_TRUNC)) {
-                       if (vp->v_type == VDIR) {
-                               error = EISDIR;
+       /* 
+        * By this point, nameidone() is called, dvp iocount is dropped,
+        * and dvp pointer is cleared.
+        */
+       if (ndp->ni_dvp != NULLVP) {
+               panic("Haven't cleaned up adequately in vn_open_auth()");
+       }
+
+       /*
+        * Expect to use this code for filesystems without compound VNOPs, for the root 
+        * of a filesystem, which can't be "looked up" in the sense of VNOP_LOOKUP(),
+        * and for shadow files, which do not live on the same filesystems as their "parents."
+        */
+       if (need_vnop_open) {
+               if (batched && !vnode_isvroot(vp) && !vnode_isnamedstream(vp)) {
+                       panic("Why am I trying to use VNOP_OPEN() on anything other than the root or a named stream?");
+               }
+
+               if (!did_create) {
+                       error = vn_authorize_open_existing(vp, &ndp->ni_cnd, fmode, ctx, NULL);
+                       if (error) {
                                goto bad;
                        }
-                       if ((error = vn_writechk(vp)) ||
-                           (error = VOP_ACCESS(vp, VWRITE, cred, p)))
+               }
+
+#if CONFIG_PROTECT
+               /* 
+                * Perform any content protection access checks prior to calling 
+                * into the filesystem, if the raw encrypted mode was not 
+                * requested.  
+                * 
+                * If the va_dataprotect_flags are NOT active, or if they are,
+                * but they do not have the VA_DP_RAWENCRYPTED bit set, then we need 
+                * to perform the checks.
+                */
+               if (!(VATTR_IS_ACTIVE (vap, va_dataprotect_flags)) ||
+                               ((vap->va_dataprotect_flags & VA_DP_RAWENCRYPTED) == 0)) {
+                       error = cp_handle_open (vp, fmode);     
+                       if (error) {
                                goto bad;
+                       }
                }
-       }
-       if (fmode & O_TRUNC) {
-               VOP_UNLOCK(vp, 0, p);                           /* XXX */
-               VOP_LEASE(vp, p, cred, LEASE_WRITE);
-               (void)vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, p);  /* XXX */
-               VATTR_NULL(vap);
-               vap->va_size = 0;
-               if (error = VOP_SETATTR(vp, vap, cred, p))
+#endif
+
+               error = VNOP_OPEN(vp, fmode, ctx);
+               if (error) {
                        goto bad;
+               }
+               need_vnop_open = FALSE;
+       }
+
+       // if the vnode is tagged VOPENEVT and the current process
+       // has the P_CHECKOPENEVT flag set, then we or in the O_EVTONLY
+       // flag to the open mode so that this open won't count against
+       // the vnode when carbon delete() does a vnode_isinuse() to see
+       // if a file is currently in use.  this allows spotlight
+       // importers to not interfere with carbon apps that depend on
+       // the no-delete-if-busy semantics of carbon delete().
+       //
+       if (!did_create && (vp->v_flag & VOPENEVT) && (current_proc()->p_flag & P_CHECKOPENEVT)) {
+               fmode |= O_EVTONLY;
        }
 
-       if (error = VOP_OPEN(vp, fmode, cred, p)) {
+       /*
+        * Grab reference, etc.
+        */
+       error = vn_open_auth_finish(vp, fmode, ctx);
+       if (error) {
+               ref_failed = TRUE;
                goto bad;
        }
 
-       if (fmode & FWRITE)
-               if (++vp->v_writecount <= 0)
-                       panic("vn_open: v_writecount");
+       /* Compound VNOP open is responsible for doing the truncate */
+       if (batched || did_create) 
+               fmode &= ~O_TRUNC;
+
+       *fmodep = fmode;
        return (0);
+
 bad:
-       VOP_UNLOCK(vp, 0, p);
-       if (didhold)
-               ubc_rele(vp);
-       vrele(vp);
+       /* Opened either explicitly or by a batched create */
+       if (!need_vnop_open) {
+               VNOP_CLOSE(vp, fmode, ctx);
+       }
+
+       ndp->ni_vp = NULL;
+       if (vp) {
+#if NAMEDRSRCFORK
+               /* Aggressively recycle shadow files if we error'd out during open() */
+               if ((vnode_isnamedstream(vp)) &&
+                       (vp->v_parent != NULLVP) && 
+                       (vnode_isshadow(vp))) {
+                               vnode_recycle(vp);
+               }
+#endif
+               vnode_put(vp);
+               /*
+                * Check for a race against unlink.  We had a vnode
+                * but according to vnode_authorize or VNOP_OPEN it
+                * no longer exists.
+                *
+                * EREDRIVEOPEN: means that we were hit by the tty allocation race.
+                */
+               if (((error == ENOENT) && (*fmodep & O_CREAT)) || (error == EREDRIVEOPEN) || ref_failed) {
+                       goto again;
+               }
+       }
+
+out:
        return (error);
 }
 
+#if vn_access_DEPRECATED
 /*
- * Check for write permissions on the specified vnode.
- * Prototype text segments cannot be written.
+ * Authorize an action against a vnode.  This has been the canonical way to
+ * ensure that the credential/process/etc. referenced by a vfs_context
+ * is granted the rights called out in 'mode' against the vnode 'vp'.
+ *
+ * Unfortunately, the use of VREAD/VWRITE/VEXEC makes it very difficult
+ * to add support for more rights.  As such, this interface will be deprecated
+ * and callers will use vnode_authorize instead.
  */
 int
-vn_writechk(vp)
-       register struct vnode *vp;
+vn_access(vnode_t vp, int mode, vfs_context_t context)
 {
-
-       /*
-        * If there's shared text associated with
-        * the vnode, try to free it up once.  If
-        * we fail, we can't allow writing.
-        */
-#if 0
-       /* XXXXX Not sure we need this */
-       if (vp->v_flag & VTEXT)
-               return (ETXTBSY);
-#endif /* 0 */
-       return (0);
+       kauth_action_t  action;
+  
+       action = 0;
+       if (mode & VREAD)
+               action |= KAUTH_VNODE_READ_DATA;
+       if (mode & VWRITE)
+               action |= KAUTH_VNODE_WRITE_DATA;
+       if (mode & VEXEC)
+               action |= KAUTH_VNODE_EXECUTE;
+  
+       return(vnode_authorize(vp, NULL, action, context));
 }
+#endif /* vn_access_DEPRECATED */
 
 /*
  * Vnode close call
  */
 int
-vn_close(vp, flags, cred, p)
-       register struct vnode *vp;
-       int flags;
-       struct ucred *cred;
-       struct proc *p;
+vn_close(struct vnode *vp, int flags, vfs_context_t ctx)
 {
        int error;
 
-       if (flags & FWRITE)
-               vp->v_writecount--;
-       error = VOP_CLOSE(vp, flags, cred, p);
-       ubc_rele(vp);
-       vrele(vp);
+#if NAMEDRSRCFORK
+       /* Sync data from resource fork shadow file if needed. */
+       if ((vp->v_flag & VISNAMEDSTREAM) && 
+           (vp->v_parent != NULLVP) &&
+           vnode_isshadow(vp)) {
+               if (flags & FWASWRITTEN) {
+                       (void) vnode_flushnamedstream(vp->v_parent, vp, ctx);
+               }
+       }
+#endif
+       
+       /* work around for foxhound */
+       if (vnode_isspec(vp))
+               (void)vnode_rele_ext(vp, flags, 0);
+
+       error = VNOP_CLOSE(vp, flags, ctx);
+
+#if CONFIG_FSE
+       if (flags & FWASWRITTEN) {
+               if (need_fsevent(FSE_CONTENT_MODIFIED, vp)) {
+                       add_fsevent(FSE_CONTENT_MODIFIED, ctx,
+                                   FSE_ARG_VNODE, vp,
+                                   FSE_ARG_DONE);
+               }
+       }
+#endif
+
+       if (!vnode_isspec(vp))
+               (void)vnode_rele_ext(vp, flags, 0);
+       
        return (error);
 }
 
+static int
+vn_read_swapfile(
+       struct vnode    *vp,
+       uio_t           uio)
+{
+       int     error;
+       off_t   swap_count, this_count;
+       off_t   file_end, read_end;
+       off_t   prev_resid;
+       char    *my_swap_page;
+
+       /*
+        * Reading from a swap file will get you zeroes.
+        */
+
+       my_swap_page = NULL;
+       error = 0;
+       swap_count = uio_resid(uio);
+
+       file_end = ubc_getsize(vp);
+       read_end = uio->uio_offset + uio_resid(uio);
+       if (uio->uio_offset >= file_end) {
+               /* uio starts after end of file: nothing to read */
+               swap_count = 0;
+       } else if (read_end > file_end) {
+               /* uio extends beyond end of file: stop before that */
+               swap_count -= (read_end - file_end);
+       }
+
+       while (swap_count > 0) {
+               if (my_swap_page == NULL) {
+                       MALLOC(my_swap_page, char *, PAGE_SIZE,
+                              M_TEMP, M_WAITOK);
+                       memset(my_swap_page, '\0', PAGE_SIZE);
+                       /* add an end-of-line to keep line counters happy */
+                       my_swap_page[PAGE_SIZE-1] = '\n';
+               }
+               this_count = swap_count;
+               if (this_count > PAGE_SIZE) {
+                       this_count = PAGE_SIZE;
+               }
+
+               prev_resid = uio_resid(uio);
+               error = uiomove((caddr_t) my_swap_page,
+                               this_count,
+                               uio);
+               if (error) {
+                       break;
+               }
+               swap_count -= (prev_resid - uio_resid(uio));
+       }
+       if (my_swap_page != NULL) {
+               FREE(my_swap_page, M_TEMP);
+               my_swap_page = NULL;
+       }
+
+       return error;
+}
 /*
  * Package up an I/O request on a vnode into a uio and do it.
  */
 int
-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;
+vn_rdwr(
+       enum uio_rw rw,
+       struct vnode *vp,
+       caddr_t base,
+       int len,
+       off_t offset,
+       enum uio_seg segflg,
+       int ioflg,
+       kauth_cred_t cred,
+       int *aresid,
+       proc_t p)
 {
-       struct uio auio;
-       struct iovec aiov;
+       int64_t resid;
+       int result;
+       
+       result = vn_rdwr_64(rw,
+                       vp,
+                       (uint64_t)(uintptr_t)base,
+                       (int64_t)len,
+                       offset,
+                       segflg,
+                       ioflg,
+                       cred,
+                       &resid,
+                       p);
+
+       /* "resid" should be bounded above by "len," which is an int */
+       if (aresid != NULL) {
+               *aresid = resid;
+       }
+
+       return result;
+}
+
+
+int
+vn_rdwr_64(
+       enum uio_rw rw,
+       struct vnode *vp,
+       uint64_t base,
+       int64_t len,
+       off_t offset,
+       enum uio_seg segflg,
+       int ioflg,
+       kauth_cred_t cred,
+       int64_t *aresid,
+       proc_t p)
+{
+       uio_t auio;
+       int spacetype;
+       struct vfs_context context;
        int error=0;
+       char uio_buf[ UIO_SIZEOF(1) ];
+
+       context.vc_thread = current_thread();
+       context.vc_ucred = cred;
+
+       if (UIO_SEG_IS_USER_SPACE(segflg)) {
+               spacetype = proc_is64bit(p) ? UIO_USERSPACE64 : UIO_USERSPACE32;
+       }
+       else {
+               spacetype = UIO_SYSSPACE;
+       }
+       auio = uio_createwithbuffer(1, offset, spacetype, rw, 
+                                                                 &uio_buf[0], sizeof(uio_buf));
+       uio_addiov(auio, base, len);
+
+#if CONFIG_MACF
+       /* XXXMAC
+        *      IO_NOAUTH should be re-examined.
+        *      Likely that mediation should be performed in caller.
+        */
+       if ((ioflg & IO_NOAUTH) == 0) {
+       /* passed cred is fp->f_cred */
+               if (rw == UIO_READ)
+                       error = mac_vnode_check_read(&context, cred, vp);
+               else
+                       error = mac_vnode_check_write(&context, cred, vp);
+       }
+#endif
 
-        /* FIXME XXX */
-       if ((ioflg & IO_NODELOCKED) == 0)
-               (void)vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, p);
-       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 (error == 0) {
+               if (rw == UIO_READ) {
+                       if (vnode_isswap(vp)) {
+                               error = vn_read_swapfile(vp, auio);
+                       } else {
+                               error = VNOP_READ(vp, auio, ioflg, &context);
+                       }
+               } else {
+                       error = VNOP_WRITE(vp, auio, ioflg, &context);
+               }
+       }
 
        if (aresid)
-               *aresid = auio.uio_resid;
+               *aresid = uio_resid(auio);
        else
-               if (auio.uio_resid && error == 0)
+               if (uio_resid(auio) && error == 0)
                        error = EIO;
-       if ((ioflg & IO_NODELOCKED) == 0)
-               VOP_UNLOCK(vp, 0, p);
        return (error);
 }
 
@@ -312,81 +882,52 @@ vn_rdwr(rw, vp, base, len, offset, segflg, ioflg, cred, aresid, p)
  * File table vnode read routine.
  */
 static int
-vn_read(fp, uio, cred, flags, p)
-       struct file *fp;
-       struct uio *uio;
-       struct ucred *cred;
-       int flags;
-       struct proc *p;
+vn_read(struct fileproc *fp, struct uio *uio, int flags, vfs_context_t ctx)
 {
        struct vnode *vp;
        int error, ioflag;
        off_t count;
 
-       if (p != uio->uio_procp)
-               panic("vn_read: uio_procp does not match p");
+       vp = (struct vnode *)fp->f_fglob->fg_data;
+       if ( (error = vnode_getwithref(vp)) ) {
+               return(error);
+       }
 
-       vp = (struct vnode *)fp->f_data;
-       ioflag = 0;
-       if (fp->f_flag & FNONBLOCK)
-               ioflag |= IO_NDELAY;
-       VOP_LEASE(vp, p, cred, LEASE_READ);
-       error = vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, p);
-       if (error)
+#if CONFIG_MACF
+       error = mac_vnode_check_read(ctx, vfs_context_ucred(ctx), vp);
+       if (error) {
+               (void)vnode_put(vp);
                return (error);
+       }
+#endif
+
+       /* This signals to VNOP handlers that this read came from a file table read */
+       ioflag = IO_SYSCALL_DISPATCH;
+
+       if (fp->f_fglob->fg_flag & FNONBLOCK)
+               ioflag |= IO_NDELAY;
+       if ((fp->f_fglob->fg_flag & FNOCACHE) || vnode_isnocache(vp))
+           ioflag |= IO_NOCACHE;
+       if (fp->f_fglob->fg_flag & FENCRYPTED) {
+               ioflag |= IO_ENCRYPTED;
+       }
+       if (fp->f_fglob->fg_flag & FNORDAHEAD)
+           ioflag |= IO_RAOFF;
+
        if ((flags & FOF_OFFSET) == 0)
-               uio->uio_offset = fp->f_offset;
-       count = uio->uio_resid;
-
-       if(UBCINFOEXISTS(vp)) {
-               memory_object_t pager;
-               struct iovec    *iov;
-               off_t           file_off;
-               kern_return_t   kr = KERN_SUCCESS;
-               kern_return_t   ret = KERN_SUCCESS;
-               int             count;
-
-               pager = (memory_object_t)ubc_getpager(vp);
-               file_off = uio->uio_offset;
-               iov = uio->uio_iov;
-               count = uio->uio_iovcnt;
-               while(count) {
-                       kr = vm_conflict_check(current_map(), 
-                               (vm_offset_t)iov->iov_base, iov->iov_len, 
-                               pager, file_off);
-                       if(kr == KERN_ALREADY_WAITING) {
-                               if((count != uio->uio_iovcnt) &&
-                                  (ret != KERN_ALREADY_WAITING)) {
-                                       error = EINVAL;
-                                       goto done;
-                               }
-                               ret = KERN_ALREADY_WAITING;
-                       } else if (kr != KERN_SUCCESS) {
-                               error = EINVAL;
-                               goto done;
-                       }
-                       if(kr != ret) {
-                               error = EINVAL;
-                               goto done;
-                       }
-                       file_off += iov->iov_len;
-                       iov++;
-                       count--;
-               }
-               if(ret == KERN_ALREADY_WAITING) {
-                       uio->uio_resid = 0;
-                       if ((flags & FOF_OFFSET) == 0)
-                               fp->f_offset += 
-                                       count - uio->uio_resid;
-                       error = 0;
-                       goto done;
-               }
+               uio->uio_offset = fp->f_fglob->fg_offset;
+       count = uio_resid(uio);
+
+       if (vnode_isswap(vp)) {
+               /* special case for swap files */
+               error = vn_read_swapfile(vp, uio);
+       } else {
+               error = VNOP_READ(vp, uio, ioflag, ctx);
        }
-       error = VOP_READ(vp, uio, ioflag, cred);
        if ((flags & FOF_OFFSET) == 0)
-               fp->f_offset += count - uio->uio_resid;
-done:
-       VOP_UNLOCK(vp, 0, p);
+               fp->f_fglob->fg_offset += count - uio_resid(uio);
+
+       (void)vnode_put(vp);
        return (error);
 }
 
@@ -395,129 +936,205 @@ done:
  * File table vnode write routine.
  */
 static int
-vn_write(fp, uio, cred, flags, p)
-       struct file *fp;
-       struct uio *uio;
-       struct ucred *cred;
-       int flags;
-       struct proc *p;
+vn_write(struct fileproc *fp, struct uio *uio, int flags, vfs_context_t ctx)
 {
        struct vnode *vp;
        int error, ioflag;
        off_t count;
+       int clippedsize = 0;
+       int partialwrite=0;
+       int residcount, oldcount;
+       proc_t p = vfs_context_proc(ctx);
 
-       if (p != uio->uio_procp)
-               panic("vn_write: uio_procp does not match p");
+       count = 0;
+       vp = (struct vnode *)fp->f_fglob->fg_data;
+       if ( (error = vnode_getwithref(vp)) ) {
+               return(error);
+       }
 
-       vp = (struct vnode *)fp->f_data;
-       ioflag = IO_UNIT;
-       if (vp->v_type == VREG)
-               bwillwrite();
-       if (vp->v_type == VREG && (fp->f_flag & O_APPEND))
+#if CONFIG_MACF
+       error = mac_vnode_check_write(ctx, vfs_context_ucred(ctx), vp);
+       if (error) {
+               (void)vnode_put(vp);
+               return (error);
+       }
+#endif
+
+       /* 
+        * IO_SYSCALL_DISPATCH signals to VNOP handlers that this write originated 
+        * from a file table write.
+        */
+       ioflag = (IO_UNIT | IO_SYSCALL_DISPATCH);
+
+       if (vp->v_type == VREG && (fp->f_fglob->fg_flag & O_APPEND))
                ioflag |= IO_APPEND;
-       if (fp->f_flag & FNONBLOCK)
+       if (fp->f_fglob->fg_flag & FNONBLOCK)
                ioflag |= IO_NDELAY;
-       if ((fp->f_flag & O_FSYNC) ||
-               (vp->v_mount && (vp->v_mount->mnt_flag & MNT_SYNCHRONOUS)))
+       if ((fp->f_fglob->fg_flag & FNOCACHE) || vnode_isnocache(vp))
+               ioflag |= IO_NOCACHE;
+       if (fp->f_fglob->fg_flag & FNODIRECT)
+               ioflag |= IO_NODIRECT;
+       if (fp->f_fglob->fg_flag & FSINGLE_WRITER)
+               ioflag |= IO_SINGLE_WRITER;
+
+       /*
+        * Treat synchronous mounts and O_FSYNC on the fd as equivalent.
+        *
+        * XXX We treat O_DSYNC as O_FSYNC for now, since we can not delay
+        * XXX the non-essential metadata without some additional VFS work;
+        * XXX the intent at this point is to plumb the interface for it.
+        */
+       if ((fp->f_fglob->fg_flag & (O_FSYNC|O_DSYNC)) ||
+               (vp->v_mount && (vp->v_mount->mnt_flag & MNT_SYNCHRONOUS))) {
                ioflag |= IO_SYNC;
-       VOP_LEASE(vp, p, cred, LEASE_WRITE);
-       error = vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, p);
-       if (error)
-               return (error);
+       }
+
        if ((flags & FOF_OFFSET) == 0) {
-               uio->uio_offset = fp->f_offset;
-               count = uio->uio_resid;
-       }
-
-       if(UBCINFOEXISTS(vp)) {
-               memory_object_t         pager;
-               struct iovec            *iov;
-               off_t                   file_off;
-               kern_return_t           kr = KERN_SUCCESS;
-               kern_return_t           ret = KERN_SUCCESS;
-               int                     count;
-
-               pager = (memory_object_t)ubc_getpager(vp);
-               file_off = uio->uio_offset;
-               iov = uio->uio_iov;
-               count = uio->uio_iovcnt;
-               while(count) {
-                       kr = vm_conflict_check(current_map(), 
-                               (vm_offset_t)iov->iov_base, 
-                               iov->iov_len, pager, file_off);
-                       if(kr == KERN_ALREADY_WAITING) {
-                               if((count != uio->uio_iovcnt) &&
-                                  (ret != KERN_ALREADY_WAITING)) {
-                                       error = EINVAL;
-                                       goto done;
-                               }
-                               ret = KERN_ALREADY_WAITING;
-                       } else if (kr != KERN_SUCCESS) {
-                               error = EINVAL;
-                               goto done;
-                       }
-                       if(kr != ret) {
-                               error = EINVAL;
-                               goto done;
-                       }
-                       file_off += iov->iov_len;
-                       iov++;
-                       count--;
+               uio->uio_offset = fp->f_fglob->fg_offset;
+               count = uio_resid(uio);
+       }
+       if (((flags & FOF_OFFSET) == 0) &&
+               vfs_context_proc(ctx) && (vp->v_type == VREG) &&
+            (((rlim_t)(uio->uio_offset + uio_resid(uio)) > p->p_rlimit[RLIMIT_FSIZE].rlim_cur) ||
+             ((rlim_t)uio_resid(uio) > (p->p_rlimit[RLIMIT_FSIZE].rlim_cur - uio->uio_offset)))) {
+               /*
+                * If the requested residual would cause us to go past the
+                * administrative limit, then we need to adjust the residual
+                * down to cause fewer bytes than requested to be written.  If
+                * we can't do that (e.g. the residual is already 1 byte),
+                * then we fail the write with EFBIG.
+                */
+               residcount = uio_resid(uio);
+               if ((rlim_t)(uio->uio_offset + uio_resid(uio)) > p->p_rlimit[RLIMIT_FSIZE].rlim_cur) {
+                       clippedsize =  (uio->uio_offset + uio_resid(uio)) - p->p_rlimit[RLIMIT_FSIZE].rlim_cur;
+               } else if ((rlim_t)uio_resid(uio) > (p->p_rlimit[RLIMIT_FSIZE].rlim_cur - uio->uio_offset)) {
+                       clippedsize = (p->p_rlimit[RLIMIT_FSIZE].rlim_cur - uio->uio_offset);
+               }
+               if (clippedsize >= residcount) {
+                       psignal(p, SIGXFSZ);
+                       vnode_put(vp);
+                       return (EFBIG);
                }
-               if(ret == KERN_ALREADY_WAITING) {
-                       uio->uio_resid = 0;
-                       if ((flags & FOF_OFFSET) == 0)
-                               fp->f_offset += 
-                                       count - uio->uio_resid;
-                       error = 0;
-                       goto done;
+               partialwrite = 1;
+               uio_setresid(uio, residcount-clippedsize);
+       }
+       if ((flags & FOF_OFFSET) != 0) {
+               /* for pwrite, append should  be ignored */
+               ioflag &= ~IO_APPEND;
+               if (p && (vp->v_type == VREG) &&
+               ((rlim_t)uio->uio_offset  >= p->p_rlimit[RLIMIT_FSIZE].rlim_cur)) {
+               psignal(p, SIGXFSZ);
+               vnode_put(vp);
+               return (EFBIG);
+       }
+               if (p && (vp->v_type == VREG) &&
+                       ((rlim_t)(uio->uio_offset + uio_resid(uio)) > p->p_rlimit[RLIMIT_FSIZE].rlim_cur)) {
+                       //Debugger("vn_bwrite:overstepping the bounds");
+                       residcount = uio_resid(uio);
+                       clippedsize =  (uio->uio_offset + uio_resid(uio)) - p->p_rlimit[RLIMIT_FSIZE].rlim_cur;
+                       partialwrite = 1;
+                       uio_setresid(uio, residcount-clippedsize);
                }
        }
-       error = VOP_WRITE(vp, uio, ioflag, cred);
+
+       error = VNOP_WRITE(vp, uio, ioflag, ctx);
+
+       if (partialwrite) {
+               oldcount = uio_resid(uio);
+               uio_setresid(uio, oldcount + clippedsize);
+       }
 
        if ((flags & FOF_OFFSET) == 0) {
                if (ioflag & IO_APPEND)
-                       fp->f_offset = uio->uio_offset;
+                       fp->f_fglob->fg_offset = uio->uio_offset;
                else
-                       fp->f_offset += count - uio->uio_resid;
+                       fp->f_fglob->fg_offset += count - uio_resid(uio);
        }
 
        /*
         * Set the credentials on successful writes
         */
        if ((error == 0) && (vp->v_tag == VT_NFS) && (UBCINFOEXISTS(vp))) {
-               ubc_setcred(vp, p);
+               /* 
+                * When called from aio subsystem, we only have the proc from
+                * which to get the credential, at this point, so use that
+                * instead.  This means aio functions are incompatible with
+                * per-thread credentials (aio operations are proxied).  We
+                * can't easily correct the aio vs. settid race in this case
+                * anyway, so we disallow it.
+                */
+               if ((flags & FOF_PCRED) == 0) {
+                       ubc_setthreadcred(vp, p, current_thread());
+               } else {
+                       ubc_setcred(vp, p);
+               }
        }
-
-done:
-       VOP_UNLOCK(vp, 0, p);
+       (void)vnode_put(vp);
        return (error);
 }
 
 /*
  * File table vnode stat routine.
+ *
+ * Returns:    0                       Success
+ *             EBADF
+ *             ENOMEM
+ *     vnode_getattr:???
  */
 int
-vn_stat(vp, sb, p)
-       struct vnode *vp;
-       register struct stat *sb;
-       struct proc *p;
+vn_stat_noauth(struct vnode *vp, void *sbptr, kauth_filesec_t *xsec, int isstat64, vfs_context_t ctx)
 {
-       struct vattr vattr;
-       register struct vattr *vap;
+       struct vnode_attr va;
        int error;
        u_short mode;
+       kauth_filesec_t fsec;
+       struct stat *sb = (struct stat *)0;     /* warning avoidance ; protected by isstat64 */
+       struct stat64 * sb64 = (struct stat64 *)0;  /* warning avoidance ; protected by isstat64 */
 
-       vap = &vattr;
-       error = VOP_GETATTR(vp, vap, p->p_ucred, p);
+       if (isstat64 != 0)
+               sb64 = (struct stat64 *)sbptr;
+       else
+               sb = (struct stat *)sbptr;
+       memset(&va, 0, sizeof(va));
+       VATTR_INIT(&va);
+       VATTR_WANTED(&va, va_fsid);
+       VATTR_WANTED(&va, va_fileid);
+       VATTR_WANTED(&va, va_mode);
+       VATTR_WANTED(&va, va_type);
+       VATTR_WANTED(&va, va_nlink);
+       VATTR_WANTED(&va, va_uid);
+       VATTR_WANTED(&va, va_gid);
+       VATTR_WANTED(&va, va_rdev);
+       VATTR_WANTED(&va, va_data_size);
+       VATTR_WANTED(&va, va_access_time);
+       VATTR_WANTED(&va, va_modify_time);
+       VATTR_WANTED(&va, va_change_time);
+       VATTR_WANTED(&va, va_create_time);
+       VATTR_WANTED(&va, va_flags);
+       VATTR_WANTED(&va, va_gen);
+       VATTR_WANTED(&va, va_iosize);
+       /* lower layers will synthesise va_total_alloc from va_data_size if required */
+       VATTR_WANTED(&va, va_total_alloc);
+       if (xsec != NULL) {
+               VATTR_WANTED(&va, va_uuuid);
+               VATTR_WANTED(&va, va_guuid);
+               VATTR_WANTED(&va, va_acl);
+       }
+       error = vnode_getattr(vp, &va, ctx);
        if (error)
-               return (error);
+               goto out;
        /*
         * Copy from vattr table
         */
-       sb->st_dev = vap->va_fsid;
-       sb->st_ino = vap->va_fileid;
-       mode = vap->va_mode;
+       if (isstat64 != 0) {
+               sb64->st_dev = va.va_fsid;
+               sb64->st_ino = (ino64_t)va.va_fileid;
+
+       } else {
+               sb->st_dev = va.va_fsid;
+               sb->st_ino = (ino_t)va.va_fileid;
+       }
+       mode = va.va_mode;
        switch (vp->v_type) {
        case VREG:
                mode |= S_IFREG;
@@ -541,150 +1158,583 @@ vn_stat(vp, sb, p)
                mode |= S_IFIFO;
                break;
        default:
-               return (EBADF);
+               error = EBADF;
+               goto out;
        };
-       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;
-       sb->st_mtimespec = vap->va_mtime;
-       sb->st_ctimespec = vap->va_ctime;
-       sb->st_blksize = vap->va_blocksize;
-       sb->st_flags = vap->va_flags;
+       if (isstat64 != 0) {
+               sb64->st_mode = mode;
+               sb64->st_nlink = VATTR_IS_SUPPORTED(&va, va_nlink) ? (u_int16_t)va.va_nlink : 1;
+               sb64->st_uid = va.va_uid;
+               sb64->st_gid = va.va_gid;
+               sb64->st_rdev = va.va_rdev;
+               sb64->st_size = va.va_data_size;
+               sb64->st_atimespec = va.va_access_time;
+               sb64->st_mtimespec = va.va_modify_time;
+               sb64->st_ctimespec = va.va_change_time;
+               sb64->st_birthtimespec = 
+                               VATTR_IS_SUPPORTED(&va, va_create_time) ? va.va_create_time : va.va_change_time;
+               sb64->st_blksize = va.va_iosize;
+               sb64->st_flags = va.va_flags;
+               sb64->st_blocks = roundup(va.va_total_alloc, 512) / 512;
+       } else {
+               sb->st_mode = mode;
+               sb->st_nlink = VATTR_IS_SUPPORTED(&va, va_nlink) ? (u_int16_t)va.va_nlink : 1;
+               sb->st_uid = va.va_uid;
+               sb->st_gid = va.va_gid;
+               sb->st_rdev = va.va_rdev;
+               sb->st_size = va.va_data_size;
+               sb->st_atimespec = va.va_access_time;
+               sb->st_mtimespec = va.va_modify_time;
+               sb->st_ctimespec = va.va_change_time;
+               sb->st_blksize = va.va_iosize;
+               sb->st_flags = va.va_flags;
+               sb->st_blocks = roundup(va.va_total_alloc, 512) / 512;
+       }
+
+       /* if we're interested in extended security data and we got an ACL */
+       if (xsec != NULL) {
+               if (!VATTR_IS_SUPPORTED(&va, va_acl) &&
+                   !VATTR_IS_SUPPORTED(&va, va_uuuid) &&
+                   !VATTR_IS_SUPPORTED(&va, va_guuid)) {
+                       *xsec = KAUTH_FILESEC_NONE;
+               } else {
+               
+                       if (VATTR_IS_SUPPORTED(&va, va_acl) && (va.va_acl != NULL)) {
+                               fsec = kauth_filesec_alloc(va.va_acl->acl_entrycount);
+                       } else {
+                               fsec = kauth_filesec_alloc(0);
+                       }
+                       if (fsec == NULL) {
+                               error = ENOMEM;
+                               goto out;
+                       }
+                       fsec->fsec_magic = KAUTH_FILESEC_MAGIC;
+                       if (VATTR_IS_SUPPORTED(&va, va_uuuid)) {
+                               fsec->fsec_owner = va.va_uuuid;
+                       } else {
+                               fsec->fsec_owner = kauth_null_guid;
+                       }
+                       if (VATTR_IS_SUPPORTED(&va, va_guuid)) {
+                               fsec->fsec_group = va.va_guuid;
+                       } else {
+                               fsec->fsec_group = kauth_null_guid;
+                       }
+                       if (VATTR_IS_SUPPORTED(&va, va_acl) && (va.va_acl != NULL)) {
+                               bcopy(va.va_acl, &(fsec->fsec_acl), KAUTH_ACL_COPYSIZE(va.va_acl));
+                       } else {
+                               fsec->fsec_acl.acl_entrycount = KAUTH_FILESEC_NOACL;
+                       }
+                       *xsec = fsec;
+               }
+       }
+       
        /* Do not give the generation number out to unpriviledged users */
-       if (suser(p->p_ucred, &p->p_acflag))
-               sb->st_gen = 0; 
-       else
-               sb->st_gen = vap->va_gen;
-       sb->st_blocks = vap->va_bytes / S_BLKSIZE;
-       return (0);
+       if (va.va_gen && !vfs_context_issuser(ctx)) {
+               if (isstat64 != 0)
+                       sb64->st_gen = 0; 
+               else
+                       sb->st_gen = 0; 
+       } else {
+               if (isstat64 != 0)
+                       sb64->st_gen = va.va_gen; 
+               else
+                       sb->st_gen = va.va_gen;
+       }
+
+       error = 0;
+out:
+       if (VATTR_IS_SUPPORTED(&va, va_acl) && va.va_acl != NULL)
+               kauth_acl_free(va.va_acl);
+       return (error);
 }
 
+int
+vn_stat(struct vnode *vp, void *sb, kauth_filesec_t *xsec, int isstat64, vfs_context_t ctx)
+{
+       int error;
+
+#if CONFIG_MACF
+       error = mac_vnode_check_stat(ctx, NOCRED, vp);
+       if (error)
+               return (error);
+#endif
+
+       /* authorize */
+       if ((error = vnode_authorize(vp, NULL, KAUTH_VNODE_READ_ATTRIBUTES | KAUTH_VNODE_READ_SECURITY, ctx)) != 0)
+               return(error);
+
+       /* actual stat */
+       return(vn_stat_noauth(vp, sb, xsec, isstat64, ctx));
+}
+
+
 /*
  * File table vnode ioctl routine.
  */
 static int
-vn_ioctl(fp, com, data, p)
-       struct file *fp;
-       u_long com;
-       caddr_t data;
-       struct proc *p;
+vn_ioctl(struct fileproc *fp, u_long com, caddr_t data, vfs_context_t ctx)
 {
-       register struct vnode *vp = ((struct vnode *)fp->f_data);
-       struct vattr vattr;
+       struct vnode *vp = ((struct vnode *)fp->f_fglob->fg_data);
+       off_t file_size;
        int error;
        struct vnode *ttyvp;
+       int funnel_state;
+       struct session * sessp;
        
-       switch (vp->v_type) {
+       if ( (error = vnode_getwithref(vp)) ) {
+               return(error);
+       }
+
+#if CONFIG_MACF
+       error = mac_vnode_check_ioctl(ctx, vp, com);
+       if (error)
+               goto out;
+#endif
 
+       switch (vp->v_type) {
        case VREG:
        case VDIR:
                if (com == FIONREAD) {
-                       if (error = VOP_GETATTR(vp, &vattr, p->p_ucred, p))
-                               return (error);
-                       *(int *)data = vattr.va_size - fp->f_offset;
-                       return (0);
+                       if ((error = vnode_size(vp, &file_size, ctx)) != 0)
+                               goto out;
+                       *(int *)data = file_size - fp->f_fglob->fg_offset;
+                       goto out;
+               }
+               if (com == FIONBIO || com == FIOASYNC) {        /* XXX */
+                       goto out;
                }
-               if (com == FIONBIO || com == FIOASYNC)  /* XXX */
-                       return (0);                     /* XXX */
                /* fall into ... */
 
        default:
-               return (ENOTTY);
+               error = ENOTTY;
+               goto out;
 
        case VFIFO:
        case VCHR:
        case VBLK:
 
-         /* Should not be able to set block size from user space */
-         if(com == DKIOCSETBLOCKSIZE)
-           return (EPERM);
-         
-         if (com == FIODTYPE) {
-           if (vp->v_type == VBLK) {
-             if (major(vp->v_rdev) >= nblkdev)
-               return (ENXIO);
-             *(int *)data = bdevsw[major(vp->v_rdev)].d_type;
-           } else if (vp->v_type == VCHR) {
-             if (major(vp->v_rdev) >= nchrdev)
-               return (ENXIO);
-             *(int *)data = cdevsw[major(vp->v_rdev)].d_type;
-           } else {
-             return (ENOTTY);
-           }
-           return (0);
-         }
-         error = VOP_IOCTL(vp, com, data, fp->f_flag, p->p_ucred, p);
-         if (error == 0 && com == TIOCSCTTY) {
-           VREF(vp);
-           ttyvp = p->p_session->s_ttyvp;
-           p->p_session->s_ttyvp = vp;
-           if (ttyvp)
-             vrele(ttyvp);
-         }
-         return (error);
+               /* Should not be able to set block size from user space */
+               if (com == DKIOCSETBLOCKSIZE) {
+                       error = EPERM;
+                       goto out;
+               }
+
+               if (com == FIODTYPE) {
+                       if (vp->v_type == VBLK) {
+                               if (major(vp->v_rdev) >= nblkdev) {
+                                       error = ENXIO;
+                                       goto out;
+                               }
+                               *(int *)data = D_TYPEMASK & bdevsw[major(vp->v_rdev)].d_type;
+
+                       } else if (vp->v_type == VCHR) {
+                               if (major(vp->v_rdev) >= nchrdev) {
+                                       error = ENXIO;
+                                       goto out;
+                               }
+                               *(int *)data = D_TYPEMASK & cdevsw[major(vp->v_rdev)].d_type;
+                       } else {
+                               error = ENOTTY;
+                               goto out;
+                       }
+                       goto out;
+               }
+               error = VNOP_IOCTL(vp, com, data, fp->f_fglob->fg_flag, ctx);
+
+               if (error == 0 && com == TIOCSCTTY) {
+                       error = vnode_ref_ext(vp, 0, VNODE_REF_FORCE);
+                       if (error != 0) {
+                               panic("vnode_ref_ext() failed despite VNODE_REF_FORCE?!");
+                       }
+
+                       funnel_state = thread_funnel_set(kernel_flock, TRUE);
+                       sessp = proc_session(vfs_context_proc(ctx));
+
+                       session_lock(sessp);
+                       ttyvp = sessp->s_ttyvp;
+                       sessp->s_ttyvp = vp;
+                       sessp->s_ttyvid = vnode_vid(vp);
+                       session_unlock(sessp);
+                       session_rele(sessp);
+                       thread_funnel_set(kernel_flock, funnel_state);
+
+                       if (ttyvp)
+                               vnode_rele(ttyvp);
+               }
        }
+out:
+       (void)vnode_put(vp);
+       return(error);
 }
 
 /*
  * File table vnode select routine.
  */
 static int
-vn_select(fp, which, wql, p)
-       struct file *fp;
-       int which;
-       void * wql;
-       struct proc *p;
+vn_select(struct fileproc *fp, int which, void *wql, __unused vfs_context_t ctx)
 {
+       int error;
+       struct vnode * vp = (struct vnode *)fp->f_fglob->fg_data;
+       struct vfs_context context;
+
+       if ( (error = vnode_getwithref(vp)) == 0 ) {
+               context.vc_thread = current_thread();
+               context.vc_ucred = fp->f_fglob->fg_cred;
+
+#if CONFIG_MACF
+               /*
+                * XXX We should use a per thread credential here; minimally,
+                * XXX the process credential should have a persistent
+                * XXX reference on it before being passed in here.
+                */
+               error = mac_vnode_check_select(ctx, vp, which);
+               if (error == 0)
+#endif
+               error = VNOP_SELECT(vp, which, fp->f_fglob->fg_flag, wql, ctx);
 
-       return(VOP_SELECT(((struct vnode *)fp->f_data), which, fp->f_flag,
-               fp->f_cred, wql, p));
+               (void)vnode_put(vp);
+       }
+       return(error);
+       
 }
 
 /*
- * Check that the vnode is still valid, and if so
- * acquire requested lock.
+ * File table vnode close routine.
+ */
+static int
+vn_closefile(struct fileglob *fg, vfs_context_t ctx)
+{
+       struct vnode *vp = (struct vnode *)fg->fg_data;
+       int error;
+       struct flock lf;
+
+       if ( (error = vnode_getwithref(vp)) == 0 ) {
+
+               if ((fg->fg_flag & FHASLOCK) && fg->fg_type == DTYPE_VNODE) {
+                       lf.l_whence = SEEK_SET;
+                       lf.l_start = 0;
+                       lf.l_len = 0;
+                       lf.l_type = F_UNLCK;
+
+                       (void)VNOP_ADVLOCK(vp, (caddr_t)fg, F_UNLCK, &lf, F_FLOCK, ctx);
+               }
+               error = vn_close(vp, fg->fg_flag, ctx);
+
+               (void)vnode_put(vp);
+       }
+       return(error);
+}
+
+/*
+ * Returns:    0                       Success
+ *     VNOP_PATHCONF:???
  */
 int
-vn_lock(vp, flags, p)
-       struct vnode *vp;
-       int flags;
-       struct proc *p;
+vn_pathconf(vnode_t vp, int name, int32_t *retval, vfs_context_t ctx)
+{
+       int     error = 0;
+       struct vfs_attr vfa;
+
+       switch(name) {
+       case _PC_EXTENDED_SECURITY_NP:
+               *retval = vfs_extendedsecurity(vnode_mount(vp)) ? 1 : 0;
+               break;
+       case _PC_AUTH_OPAQUE_NP:
+               *retval = vfs_authopaque(vnode_mount(vp));
+               break;
+       case _PC_2_SYMLINKS:
+               *retval = 1;    /* XXX NOTSUP on MSDOS, etc. */
+               break;
+       case _PC_ALLOC_SIZE_MIN:
+               *retval = 1;    /* XXX lie: 1 byte */
+               break;
+       case _PC_ASYNC_IO:      /* unistd.h: _POSIX_ASYNCHRONUS_IO */
+               *retval = 1;    /* [AIO] option is supported */
+               break;
+       case _PC_PRIO_IO:       /* unistd.h: _POSIX_PRIORITIZED_IO */
+               *retval = 0;    /* [PIO] option is not supported */
+               break;
+       case _PC_REC_INCR_XFER_SIZE:
+               *retval = 4096; /* XXX go from MIN to MAX 4K at a time */
+               break;
+       case _PC_REC_MIN_XFER_SIZE:
+               *retval = 4096; /* XXX recommend 4K minimum reads/writes */
+               break;
+       case _PC_REC_MAX_XFER_SIZE:
+               *retval = 65536; /* XXX recommend 64K maximum reads/writes */
+               break;
+       case _PC_REC_XFER_ALIGN:
+               *retval = 4096; /* XXX recommend page aligned buffers */
+               break;
+       case _PC_SYMLINK_MAX:
+               *retval = 255;  /* Minimum acceptable POSIX value */
+               break;
+       case _PC_SYNC_IO:       /* unistd.h: _POSIX_SYNCHRONIZED_IO */
+               *retval = 0;    /* [SIO] option is not supported */
+               break;
+       case _PC_XATTR_SIZE_BITS:
+               /* The number of bits used to store maximum extended 
+                * attribute size in bytes.  For example, if the maximum 
+                * attribute size supported by a file system is 128K, the 
+                * value returned will be 18.  However a value 18 can mean 
+                * that the maximum attribute size can be anywhere from 
+                * (256KB - 1) to 128KB.  As a special case, the resource 
+                * fork can have much larger size, and some file system 
+                * specific extended attributes can have smaller and preset 
+                * size; for example, Finder Info is always 32 bytes.
+                */
+               memset(&vfa, 0, sizeof(vfa));
+               VFSATTR_INIT(&vfa);
+               VFSATTR_WANTED(&vfa, f_capabilities);
+               if (vfs_getattr(vnode_mount(vp), &vfa, ctx) == 0 &&
+                   (VFSATTR_IS_SUPPORTED(&vfa, f_capabilities)) && 
+                   (vfa.f_capabilities.capabilities[VOL_CAPABILITIES_INTERFACES] & VOL_CAP_INT_EXTENDED_ATTR) && 
+                   (vfa.f_capabilities.valid[VOL_CAPABILITIES_INTERFACES] & VOL_CAP_INT_EXTENDED_ATTR)) {
+                       /* Supports native extended attributes */
+                       error = VNOP_PATHCONF(vp, name, retval, ctx);
+               } else {
+                       /* Number of bits used to represent the maximum size of 
+                        * extended attribute stored in an Apple Double file.
+                        */
+                       *retval = AD_XATTR_SIZE_BITS;
+               }
+               break;
+       default:
+               error = VNOP_PATHCONF(vp, name, retval, ctx);
+               break;
+       }
+
+       return (error);
+}
+
+static int
+vn_kqfilt_add(struct fileproc *fp, struct knote *kn, vfs_context_t ctx)
 {
        int error;
+       struct vnode *vp;
        
-       do {
-               if ((flags & LK_INTERLOCK) == 0)
-                       simple_lock(&vp->v_interlock);
-               if (vp->v_flag & VXLOCK) {
-                       while (vp->v_flag & VXLOCK) {
-                               vp->v_flag |= VXWANT;
-                               simple_unlock(&vp->v_interlock);
-                               (void)tsleep((caddr_t)vp, PINOD, "vn_lock", 0);
-                       }
-                       error = ENOENT;
-               } else {
-                       error = VOP_LOCK(vp, flags | LK_INTERLOCK, p);
-                       if (error == 0)
-                               return (error);
+       vp = (struct vnode *)fp->f_fglob->fg_data;
+
+       /*
+        * Don't attach a knote to a dead vnode.
+        */
+       if ((error = vget_internal(vp, 0, VNODE_NODEAD)) == 0) {
+               switch (kn->kn_filter) {
+                       case EVFILT_READ:
+                       case EVFILT_WRITE:
+                               if (vnode_isfifo(vp)) {
+                                       /* We'll only watch FIFOs that use our fifofs */
+                                       if (!(vp->v_fifoinfo && vp->v_fifoinfo->fi_readsock)) {
+                                               error = ENOTSUP;
+                                       }
+
+                               } else if (!vnode_isreg(vp)) {
+                                       if (vnode_ischr(vp) && 
+                                                       (error = spec_kqfilter(vp, kn)) == 0) {
+                                               /* claimed by a special device */
+                                               vnode_put(vp);
+                                               return 0;
+                                       }
+
+                                       error = EINVAL;
+                               }
+                               break;
+                       case EVFILT_VNODE:
+                               break;
+                       default:
+                               error = EINVAL;
                }
-               flags &= ~LK_INTERLOCK;
-       } while (flags & LK_RETRY);
+
+               if (error) {
+                       vnode_put(vp);
+                       return error;
+               }
+
+#if CONFIG_MACF
+               error = mac_vnode_check_kqfilter(ctx, fp->f_fglob->fg_cred, kn, vp);
+               if (error) {
+                       vnode_put(vp);
+                       return error;
+               }
+#endif
+
+               kn->kn_hook = (void*)vp;
+               kn->kn_hookid = vnode_vid(vp);
+               kn->kn_fop = &vnode_filtops;
+
+               vnode_lock(vp);
+               KNOTE_ATTACH(&vp->v_knotes, kn);
+               vnode_unlock(vp);
+
+               /* Ask the filesystem to provide remove notifications, but ignore failure */
+               VNOP_MONITOR(vp, 0, VNODE_MONITOR_BEGIN, (void*) kn,  ctx);
+
+               vnode_put(vp);
+       }
+
        return (error);
 }
 
+static void
+filt_vndetach(struct knote *kn)
+{
+       vfs_context_t ctx = vfs_context_current();
+       struct vnode *vp; 
+       vp = (struct vnode *)kn->kn_hook;
+       if (vnode_getwithvid(vp, kn->kn_hookid))
+               return;
+
+       vnode_lock(vp);
+       KNOTE_DETACH(&vp->v_knotes, kn);
+       vnode_unlock(vp);
+       
+       /* 
+        * Tell a (generally networked) filesystem that we're no longer watching 
+        * If the FS wants to track contexts, it should still be using the one from
+        * the VNODE_MONITOR_BEGIN.
+        */
+       VNOP_MONITOR(vp, 0, VNODE_MONITOR_END, (void*)kn, ctx);
+       vnode_put(vp);
+}
+
+
 /*
- * File table vnode close routine.
+ * Used for EVFILT_READ
+ *
+ * Takes only VFIFO or VREG. vnode is locked.  We handle the "poll" case
+ * differently than the regular case for VREG files.  If not in poll(),
+ * then we need to know current fileproc offset for VREG.
+ */
+static intptr_t
+vnode_readable_data_count(vnode_t vp, off_t current_offset, int ispoll)
+{
+       if (vnode_isfifo(vp)) {
+               int cnt;
+               int err = fifo_charcount(vp, &cnt);
+               if (err == 0) {
+                       return (intptr_t)cnt;
+               } else {
+                       return (intptr_t)0;
+               }
+       } else if (vnode_isreg(vp)) {
+               if (ispoll) {
+                       return (intptr_t)1;
+               }
+
+               off_t amount;
+               amount = vp->v_un.vu_ubcinfo->ui_size - current_offset;
+               if (amount > (off_t)INTPTR_MAX) {
+                       return INTPTR_MAX;
+               } else if (amount < (off_t)INTPTR_MIN) {
+                       return INTPTR_MIN;
+               } else {
+                       return (intptr_t)amount;
+               } 
+       } else {
+               panic("Should never have an EVFILT_READ except for reg or fifo.");
+               return 0;
+       }
+}
+
+/*
+ * Used for EVFILT_WRITE.  
+ *
+ * For regular vnodes, we can always write (1).  For named pipes,
+ * see how much space there is in the buffer.  Nothing else is covered.
+ */
+static intptr_t
+vnode_writable_space_count(vnode_t vp) 
+{
+       if (vnode_isfifo(vp)) {
+               long spc;
+               int err = fifo_freespace(vp, &spc);
+               if (err == 0) {
+                       return (intptr_t)spc;
+               } else {
+                       return (intptr_t)0;
+               }
+       } else if (vnode_isreg(vp)) {
+               return (intptr_t)1;
+       } else {
+               panic("Should never have an EVFILT_READ except for reg or fifo.");
+               return 0;
+       }
+}
+
+/* 
+ * Determine whether this knote should be active
+ * 
+ * This is kind of subtle.  
+ *     --First, notice if the vnode has been revoked: in so, override hint
+ *     --EVFILT_READ knotes are checked no matter what the hint is
+ *     --Other knotes activate based on hint.  
+ *     --If hint is revoke, set special flags and activate
  */
 static int
-vn_closefile(fp, p)
-       struct file *fp;
-       struct proc *p;
+filt_vnode(struct knote *kn, long hint)
 {
+       vnode_t vp = (struct vnode *)kn->kn_hook;
+       int activate = 0;
+       long orig_hint = hint;
+
+       if (0 == hint) {
+               vnode_lock(vp);
+
+               if (vnode_getiocount(vp, kn->kn_hookid, VNODE_NODEAD | VNODE_WITHID) != 0) {
+                       /* Is recycled */
+                       hint = NOTE_REVOKE;
+               } 
+       } else {
+               lck_mtx_assert(&vp->v_lock, LCK_MTX_ASSERT_OWNED);
+       }
+
+       /* Special handling for vnodes that are in recycle or already gone */
+       if (NOTE_REVOKE == hint) {
+               kn->kn_flags |= (EV_EOF | EV_ONESHOT);
+               activate = 1;
+
+               if ((kn->kn_filter == EVFILT_VNODE) && (kn->kn_sfflags & NOTE_REVOKE)) {
+                       kn->kn_fflags |= NOTE_REVOKE;
+               }
+       } else {
+               switch(kn->kn_filter) {
+                       case EVFILT_READ:
+                               kn->kn_data = vnode_readable_data_count(vp, kn->kn_fp->f_fglob->fg_offset, (kn->kn_flags & EV_POLL));
+
+                               if (kn->kn_data != 0) {
+                                       activate = 1;
+                               }
+                               break;
+                       case EVFILT_WRITE: 
+                               kn->kn_data = vnode_writable_space_count(vp);
+
+                               if (kn->kn_data != 0) {
+                                       activate = 1;
+                               }
+                               break;
+                       case EVFILT_VNODE:
+                               /* Check events this note matches against the hint */
+                               if (kn->kn_sfflags & hint) {
+                                       kn->kn_fflags |= hint; /* Set which event occurred */
+                               }
+                               if (kn->kn_fflags != 0) {
+                                       activate = 1;
+                               }
+                               break;
+                       default:
+                               panic("Invalid knote filter on a vnode!\n");
+               }
+       }
+
+       if (orig_hint == 0) {
+               /*
+                * Definitely need to unlock, may need to put 
+                */
+               if (hint == 0) {
+                       vnode_put_locked(vp);
+               }
+               vnode_unlock(vp);
+       }
 
-       return (vn_close(((struct vnode *)fp->f_data), fp->f_flag,
-               fp->f_cred, p));
+       return (activate);
 }