xnu-7195.101.1.tar.gz

[apple/xnu.git] / bsd / kern / kern_descrip.c
diff --git a/bsd/kern/kern_descrip.c b/bsd/kern/kern_descrip.c

index d906cf4405fb9d9235fe026666cfaf5088148fe7..8952d0220f16fa32da57d169b7f48e76201b36cb 100644 (file)
--- a/bsd/kern/kern_descrip.c
+++ b/bsd/kern/kern_descrip.c
@@ -99,49 +99,43 @@
  #include <sys/uio_internal.h>
  #include <sys/codesign.h>
  #include <sys/codedir_internal.h>
  #include <sys/uio_internal.h>
  #include <sys/codesign.h>
  #include <sys/codedir_internal.h>
-
-#include <security/audit/audit.h>
-
  #include <sys/mount_internal.h>
  #include <sys/kdebug.h>
  #include <sys/sysproto.h>
  #include <sys/pipe.h>
  #include <sys/spawn.h>
  #include <sys/cprotect.h>
  #include <sys/mount_internal.h>
  #include <sys/kdebug.h>
  #include <sys/sysproto.h>
  #include <sys/pipe.h>
  #include <sys/spawn.h>
  #include <sys/cprotect.h>
+#include <sys/ubc_internal.h>
+
  #include <kern/kern_types.h>
  #include <kern/kalloc.h>
  #include <kern/waitq.h>
  #include <kern/kern_types.h>
  #include <kern/kalloc.h>
  #include <kern/waitq.h>
-#include <libkern/OSAtomic.h>
-
-#include <sys/ubc_internal.h>
-
  #include <kern/ipc_misc.h>
  #include <kern/ipc_misc.h>
-#include <vm/vm_protos.h>
  
  
+#include <vm/vm_protos.h>
  #include <mach/mach_port.h>
  #include <mach/mach_port.h>
-#include <stdbool.h>
  
  
+#include <security/audit/audit.h>
  #if CONFIG_MACF
  #include <security/mac_framework.h>
  #endif
  
  #if CONFIG_MACF
  #include <security/mac_framework.h>
  #endif
  
+#include <stdbool.h>
+#include <os/atomic_private.h>
+#include <IOKit/IOBSD.h>
+
+#define IPC_OBJECT_COPYIN_FLAGS_ALLOW_IMMOVABLE_SEND 0x1
  kern_return_t ipc_object_copyin(ipc_space_t, mach_port_name_t,
  kern_return_t ipc_object_copyin(ipc_space_t, mach_port_name_t,
-    mach_msg_type_name_t, ipc_port_t *);
+    mach_msg_type_name_t, ipc_port_t *, mach_port_context_t, mach_msg_guard_flags_t *, uint32_t);
  void ipc_port_release_send(ipc_port_t);
  
  void ipc_port_release_send(ipc_port_t);
  
-struct psemnode;
-struct pshmnode;
-
+static void fileproc_drain(proc_t, struct fileproc *);
  static int finishdup(proc_t p,
      struct filedesc *fdp, int old, int new, int flags, int32_t *retval);
  
  static int finishdup(proc_t p,
      struct filedesc *fdp, int old, int new, int flags, int32_t *retval);
  
-int falloc_locked(proc_t p, struct fileproc **resultfp, int *resultfd, vfs_context_t ctx, int locked);
-void fg_drop(struct fileproc * fp);
-void fg_free(struct fileglob *fg);
-void fg_ref(struct fileproc * fp);
  void fileport_releasefg(struct fileglob *fg);
  
  void fileport_releasefg(struct fileglob *fg);
  
-/* flags for close_internal_locked */
+/* flags for fp_close_and_unlock */
  #define FD_DUP2RESV 1
  
  /* We don't want these exported */
  #define FD_DUP2RESV 1
  
  /* We don't want these exported */
@@ -149,12 +143,9 @@ void fileport_releasefg(struct fileglob *fg);
  __private_extern__
  int unlink1(vfs_context_t, vnode_t, user_addr_t, enum uio_seg, int);
  
  __private_extern__
  int unlink1(vfs_context_t, vnode_t, user_addr_t, enum uio_seg, int);
  
-static void _fdrelse(struct proc * p, int fd);
-
+static void fdrelse(struct proc * p, int fd);
  
  
-extern void file_lock_init(void);
-
-extern kauth_scope_t   kauth_scope_fileop;
+extern kauth_scope_t    kauth_scope_fileop;
  
  /* Conflict wait queue for when selects collide (opaque type) */
  extern struct waitq select_conflict_queue;
  
  /* Conflict wait queue for when selects collide (opaque type) */
  extern struct waitq select_conflict_queue;
@@ -171,31 +162,182 @@ extern struct waitq select_conflict_queue;
  #define APFSIOC_REVERT_TO_SNAPSHOT  _IOW('J', 1, u_int64_t)
  #endif
  
  #define APFSIOC_REVERT_TO_SNAPSHOT  _IOW('J', 1, u_int64_t)
  #endif
  
-#define f_flag f_fglob->fg_flag
-#define f_type f_fglob->fg_ops->fo_type
-#define f_msgcount f_fglob->fg_msgcount
-#define f_cred f_fglob->fg_cred
-#define f_ops f_fglob->fg_ops
-#define f_offset f_fglob->fg_offset
-#define f_data f_fglob->fg_data
+#define f_flag fp_glob->fg_flag
+#define f_type fp_glob->fg_ops->fo_type
+#define f_cred fp_glob->fg_cred
+#define f_ops fp_glob->fg_ops
+#define f_offset fp_glob->fg_offset
+#define f_data fp_glob->fg_data
  #define CHECK_ADD_OVERFLOW_INT64L(x, y) \
  #define CHECK_ADD_OVERFLOW_INT64L(x, y) \
-               (((((x) > 0) && ((y) > 0) && ((x) > LLONG_MAX - (y))) || \
-               (((x) < 0) && ((y) < 0) && ((x) < LLONG_MIN - (y)))) \
-               ? 1 : 0)
+               (((((x) > 0) && ((y) > 0) && ((x) > LLONG_MAX - (y))) || \
+               (((x) < 0) && ((y) < 0) && ((x) < LLONG_MIN - (y)))) \
+               ? 1 : 0)
+
+ZONE_DECLARE(fg_zone, "fileglob",
+    sizeof(struct fileglob), ZC_NOENCRYPT | ZC_ZFREE_CLEARMEM);
+ZONE_DECLARE(fp_zone, "fileproc",
+    sizeof(struct fileproc), ZC_NOENCRYPT | ZC_ZFREE_CLEARMEM);
+ZONE_DECLARE(fdp_zone, "filedesc",
+    sizeof(struct filedesc), ZC_NOENCRYPT | ZC_ZFREE_CLEARMEM);
+/*
+ * If you need accounting for KM_OFILETABL consider using
+ * KALLOC_HEAP_DEFINE to define a view.
+ */
+#define KM_OFILETABL KHEAP_DEFAULT
+
  /*
   * Descriptor management.
   */
  /*
   * Descriptor management.
   */
-struct fmsglist fmsghead;      /* head of list of open files */
-struct fmsglist fmsg_ithead;   /* head of list of open files */
-int nfiles;                    /* actual number of open files */
+int nfiles;                     /* actual number of open files */
+/*
+ * "uninitialized" ops -- ensure FILEGLOB_DTYPE(fg) always exists
+ */
+static const struct fileops uninitops;
+
+os_refgrp_decl(, f_refgrp, "files refcounts", NULL);
+static LCK_GRP_DECLARE(file_lck_grp, "file");
+
+#pragma mark fileglobs
+
+/*!
+ * @function fg_free
+ *
+ * @brief
+ * Free a file structure.
+ */
+static void
+fg_free(struct fileglob *fg)
+{
+       os_atomic_dec(&nfiles, relaxed);
+
+       if (fg->fg_vn_data) {
+               fg_vn_data_free(fg->fg_vn_data);
+               fg->fg_vn_data = NULL;
+       }
+
+       if (IS_VALID_CRED(fg->fg_cred)) {
+               kauth_cred_unref(&fg->fg_cred);
+       }
+       lck_mtx_destroy(&fg->fg_lock, &file_lck_grp);
+
+#if CONFIG_MACF
+       mac_file_label_destroy(fg);
+#endif
+       zfree(fg_zone, fg);
+}
+
+OS_ALWAYS_INLINE
+void
+fg_ref(proc_t p, struct fileglob *fg)
+{
+#if DEBUG || DEVELOPMENT
+       proc_fdlock_assert(p, LCK_MTX_ASSERT_OWNED);
+#else
+       (void)p;
+#endif
+       os_ref_retain_raw(&fg->fg_count, &f_refgrp);
+}
+
+void
+fg_drop_live(struct fileglob *fg)
+{
+       os_ref_release_live_raw(&fg->fg_count, &f_refgrp);
+}
+
+int
+fg_drop(proc_t p, struct fileglob *fg)
+{
+       struct vnode *vp;
+       struct vfs_context context;
+       int error = 0;
+
+       if (fg == NULL) {
+               return 0;
+       }
+
+       /* Set up context with cred stashed in fg */
+       if (p == current_proc()) {
+               context.vc_thread = current_thread();
+       } else {
+               context.vc_thread = NULL;
+       }
+       context.vc_ucred = fg->fg_cred;
+
+       /*
+        * POSIX record locking dictates that any close releases ALL
+        * locks owned by this process.  This is handled by setting
+        * a flag in the unlock to free ONLY locks obeying POSIX
+        * semantics, and not to free BSD-style file locks.
+        * If the descriptor was in a message, POSIX-style locks
+        * aren't passed with the descriptor.
+        */
+       if (p && DTYPE_VNODE == FILEGLOB_DTYPE(fg) &&
+           (p->p_ladvflag & P_LADVLOCK)) {
+               struct flock lf = {
+                       .l_whence = SEEK_SET,
+                       .l_type = F_UNLCK,
+               };
+
+               vp = (struct vnode *)fg->fg_data;
+               if ((error = vnode_getwithref(vp)) == 0) {
+                       (void)VNOP_ADVLOCK(vp, (caddr_t)p, F_UNLCK, &lf, F_POSIX, &context, NULL);
+                       (void)vnode_put(vp);
+               }
+       }
+
+       if (os_ref_release_raw(&fg->fg_count, &f_refgrp) == 0) {
+               /*
+                * Since we ensure that fg->fg_ops is always initialized,
+                * it is safe to invoke fo_close on the fg
+                */
+               error = fo_close(fg, &context);
+
+               fg_free(fg);
+       }
+
+       return error;
+}
+
+/*
+ * fg_get_vnode
+ *
+ * Description:        Return vnode associated with the file structure, if
+ *             any.  The lifetime of the returned vnode is bound to
+ *             the lifetime of the file structure.
+ *
+ * Parameters: fg                              Pointer to fileglob to
+ *                                             inspect
+ *
+ * Returns:    vnode_t
+ */
+vnode_t
+fg_get_vnode(struct fileglob *fg)
+{
+       if (FILEGLOB_DTYPE(fg) == DTYPE_VNODE) {
+               return (vnode_t)fg->fg_data;
+       } else {
+               return NULL;
+       }
+}
  
  
+bool
+fg_sendable(struct fileglob *fg)
+{
+       switch (FILEGLOB_DTYPE(fg)) {
+       case DTYPE_VNODE:
+       case DTYPE_SOCKET:
+       case DTYPE_PIPE:
+       case DTYPE_PSXSHM:
+       case DTYPE_NETPOLICY:
+               return (fg->fg_lflags & FG_CONFINED) == 0;
  
  
-lck_grp_attr_t * file_lck_grp_attr;
-lck_grp_t * file_lck_grp;
-lck_attr_t * file_lck_attr;
+       default:
+               return false;
+       }
+}
  
  
-lck_mtx_t * uipc_lock;
  
  
+#pragma mark fileprocs
  
  /*
   * check_file_seek_range
  
  /*
   * check_file_seek_range
@@ -205,9 +347,9 @@ lck_mtx_t * uipc_lock;
   * Parameters:  fl             Flock structure.
   *             cur_file_offset Current offset in the file.
   *
   * Parameters:  fl             Flock structure.
   *             cur_file_offset Current offset in the file.
   *
- * Returns:    0               on Success.
+ * Returns:     0               on Success.
   *             EOVERFLOW       on overflow.
   *             EOVERFLOW       on overflow.
- *             EINVAL          on offset less than zero.
+ *             EINVAL          on offset less than zero.
   */
  
  static int
   */
  
  static int
@@ -228,12 +370,12 @@ check_file_seek_range(struct flock *fl, off_t cur_file_offset)
                 }
                 /* Check if end marker is beyond LLONG_MAX. */
                 if ((fl->l_len > 0) && (CHECK_ADD_OVERFLOW_INT64L(fl->l_start +
                 }
                 /* Check if end marker is beyond LLONG_MAX. */
                 if ((fl->l_len > 0) && (CHECK_ADD_OVERFLOW_INT64L(fl->l_start +
-                       cur_file_offset, fl->l_len - 1))) {
+                   cur_file_offset, fl->l_len - 1))) {
                         return EOVERFLOW;
                 }
                 /* Check if the end marker is negative. */
                 if ((fl->l_len <= 0) && (fl->l_start + cur_file_offset +
                         return EOVERFLOW;
                 }
                 /* Check if the end marker is negative. */
                 if ((fl->l_len <= 0) && (fl->l_start + cur_file_offset +
-                       fl->l_len < 0)) {
+                   fl->l_len < 0)) {
                         return EINVAL;
                 }
         } else if (fl->l_whence == SEEK_SET) {
                         return EINVAL;
                 }
         } else if (fl->l_whence == SEEK_SET) {
@@ -247,7 +389,7 @@ check_file_seek_range(struct flock *fl, off_t cur_file_offset)
                         return EOVERFLOW;
                 }
                 /* Check if the end marker is negative. */
                         return EOVERFLOW;
                 }
                 /* Check if the end marker is negative. */
-               if ((fl->l_len < 0) &&  fl->l_start + fl->l_len < 0) {
+               if ((fl->l_len < 0) && fl->l_start + fl->l_len < 0) {
                         return EINVAL;
                 }
         }
                         return EINVAL;
                 }
         }
@@ -255,31 +397,29 @@ check_file_seek_range(struct flock *fl, off_t cur_file_offset)
  }
  
  
  }
  
  
-/*
- * file_lock_init
- *
- * Description:        Initialize the file lock group and the uipc and flist locks
- *
- * Parameters: (void)
- *
- * Returns:    void
- *
- * Notes:      Called at system startup from bsd_init().
- */
  void
  void
-file_lock_init(void)
+proc_dirs_lock_shared(proc_t p)
  {
  {
-       /* allocate file lock group attribute and group */
-       file_lck_grp_attr= lck_grp_attr_alloc_init();
-
-       file_lck_grp = lck_grp_alloc_init("file",  file_lck_grp_attr);
+       lck_rw_lock_shared(&p->p_dirs_lock);
+}
  
  
-       /* Allocate file lock attribute */
-       file_lck_attr = lck_attr_alloc_init();
+void
+proc_dirs_unlock_shared(proc_t p)
+{
+       lck_rw_unlock_shared(&p->p_dirs_lock);
+}
  
  
-       uipc_lock = lck_mtx_alloc_init(file_lck_grp, file_lck_attr);
+void
+proc_dirs_lock_exclusive(proc_t p)
+{
+       lck_rw_lock_exclusive(&p->p_dirs_lock);
  }
  
  }
  
+void
+proc_dirs_unlock_exclusive(proc_t p)
+{
+       lck_rw_unlock_exclusive(&p->p_dirs_lock);
+}
  
  /*
   * proc_fdlock, proc_fdlock_spin
  
  /*
   * proc_fdlock, proc_fdlock_spin
@@ -328,6 +468,56 @@ proc_fdunlock(proc_t p)
         lck_mtx_unlock(&p->p_fdmlock);
  }
  
         lck_mtx_unlock(&p->p_fdmlock);
  }
  
+struct fdt_iterator
+fdt_next(proc_t p, int fd, bool only_settled)
+{
+       struct fdt_iterator it;
+       struct filedesc *fdp = p->p_fd;
+       struct fileproc *fp;
+       int nfds = min(fdp->fd_lastfile + 1, fdp->fd_nfiles);
+
+       while (++fd < nfds) {
+               fp = fdp->fd_ofiles[fd];
+               if (fp == NULL || fp->fp_glob == NULL) {
+                       continue;
+               }
+               if (only_settled && (fdp->fd_ofileflags[fd] & UF_RESERVED)) {
+                       continue;
+               }
+               it.fdti_fd = fd;
+               it.fdti_fp = fp;
+               return it;
+       }
+
+       it.fdti_fd = nfds;
+       it.fdti_fp = NULL;
+       return it;
+}
+
+struct fdt_iterator
+fdt_prev(proc_t p, int fd, bool only_settled)
+{
+       struct fdt_iterator it;
+       struct filedesc *fdp = p->p_fd;
+       struct fileproc *fp;
+
+       while (--fd >= 0) {
+               fp = fdp->fd_ofiles[fd];
+               if (fp == NULL || fp->fp_glob == NULL) {
+                       continue;
+               }
+               if (only_settled && (fdp->fd_ofileflags[fd] & UF_RESERVED)) {
+                       continue;
+               }
+               it.fdti_fd = fd;
+               it.fdti_fp = fp;
+               return it;
+       }
+
+       it.fdti_fd = -1;
+       it.fdti_fp = NULL;
+       return it;
+}
  
  /*
   * System calls on descriptors.
  
  /*
   * System calls on descriptors.
@@ -335,7 +525,7 @@ proc_fdunlock(proc_t p)
  
  
  /*
  
  
  /*
- * getdtablesize
+ * sys_getdtablesize
   *
   * Description:        Returns the per process maximum size of the descriptor table
   *
   *
   * Description:        Returns the per process maximum size of the descriptor table
   *
@@ -348,50 +538,42 @@ proc_fdunlock(proc_t p)
   *             *retval (modified)              Size of dtable
   */
  int
   *             *retval (modified)              Size of dtable
   */
  int
-getdtablesize(proc_t p, __unused struct getdtablesize_args *uap, int32_t *retval)
+sys_getdtablesize(proc_t p, __unused struct getdtablesize_args *uap, int32_t *retval)
  {
  {
-       proc_fdlock_spin(p);
-       *retval = min((int)p->p_rlimit[RLIMIT_NOFILE].rlim_cur, maxfiles);
-       proc_fdunlock(p);
+       *retval = (int32_t)MIN(proc_limitgetcur(p, RLIMIT_NOFILE, TRUE), maxfilesperproc);
  
  
-       return (0);
+       return 0;
  }
  
  
  }
  
  
-void
+static void
  procfdtbl_reservefd(struct proc * p, int fd)
  {
         p->p_fd->fd_ofiles[fd] = NULL;
  procfdtbl_reservefd(struct proc * p, int fd)
  {
         p->p_fd->fd_ofiles[fd] = NULL;
-        p->p_fd->fd_ofileflags[fd] |= UF_RESERVED;
-}
-
-void
-procfdtbl_markclosefd(struct proc * p, int fd)
-{
-        p->p_fd->fd_ofileflags[fd] |= (UF_RESERVED | UF_CLOSING);
+       p->p_fd->fd_ofileflags[fd] |= UF_RESERVED;
  }
  
  void
  procfdtbl_releasefd(struct proc * p, int fd, struct fileproc * fp)
  {
  }
  
  void
  procfdtbl_releasefd(struct proc * p, int fd, struct fileproc * fp)
  {
-       if (fp != NULL)
-               p->p_fd->fd_ofiles[fd] = fp;
-        p->p_fd->fd_ofileflags[fd] &= ~UF_RESERVED;
+       if (fp != NULL) {
+               p->p_fd->fd_ofiles[fd] = fp;
+       }
+       p->p_fd->fd_ofileflags[fd] &= ~UF_RESERVED;
         if ((p->p_fd->fd_ofileflags[fd] & UF_RESVWAIT) == UF_RESVWAIT) {
                 p->p_fd->fd_ofileflags[fd] &= ~UF_RESVWAIT;
                 wakeup(&p->p_fd);
         }
  }
  
         if ((p->p_fd->fd_ofileflags[fd] & UF_RESVWAIT) == UF_RESVWAIT) {
                 p->p_fd->fd_ofileflags[fd] &= ~UF_RESVWAIT;
                 wakeup(&p->p_fd);
         }
  }
  
-void
+static void
  procfdtbl_waitfd(struct proc * p, int fd)
  {
  procfdtbl_waitfd(struct proc * p, int fd)
  {
-        p->p_fd->fd_ofileflags[fd] |= UF_RESVWAIT;
+       p->p_fd->fd_ofileflags[fd] |= UF_RESVWAIT;
         msleep(&p->p_fd, &p->p_fdmlock, PRIBIO, "ftbl_waitfd", NULL);
  }
  
         msleep(&p->p_fd, &p->p_fdmlock, PRIBIO, "ftbl_waitfd", NULL);
  }
  
-
-void
+static void
  procfdtbl_clearfd(struct proc * p, int fd)
  {
         int waiting;
  procfdtbl_clearfd(struct proc * p, int fd)
  {
         int waiting;
@@ -399,13 +581,13 @@ procfdtbl_clearfd(struct proc * p, int fd)
         waiting = (p->p_fd->fd_ofileflags[fd] & UF_RESVWAIT);
         p->p_fd->fd_ofiles[fd] = NULL;
         p->p_fd->fd_ofileflags[fd] = 0;
         waiting = (p->p_fd->fd_ofileflags[fd] & UF_RESVWAIT);
         p->p_fd->fd_ofiles[fd] = NULL;
         p->p_fd->fd_ofileflags[fd] = 0;
-       if ( waiting == UF_RESVWAIT) {
+       if (waiting == UF_RESVWAIT) {
                 wakeup(&p->p_fd);
         }
  }
  
  /*
                 wakeup(&p->p_fd);
         }
  }
  
  /*
- * _fdrelse
+ * fdrelse
   *
   * Description:        Inline utility function to free an fd in a filedesc
   *
   *
   * Description:        Inline utility function to free an fd in a filedesc
   *
@@ -419,24 +601,27 @@ procfdtbl_clearfd(struct proc * p, int fd)
   *             the caller
   */
  static void
   *             the caller
   */
  static void
-_fdrelse(struct proc * p, int fd)
+fdrelse(struct proc * p, int fd)
  {
         struct filedesc *fdp = p->p_fd;
         int nfd = 0;
  
  {
         struct filedesc *fdp = p->p_fd;
         int nfd = 0;
  
-       if (fd < fdp->fd_freefile)
+       if (fd < fdp->fd_freefile) {
                 fdp->fd_freefile = fd;
                 fdp->fd_freefile = fd;
+       }
  #if DIAGNOSTIC
  #if DIAGNOSTIC
-       if (fd > fdp->fd_lastfile)
-               panic("fdrelse: fd_lastfile inconsistent");
+       if (fd > fdp->fd_lastfile) {
+               panic("fdrelse: fd_lastfile inconsistent");
+       }
  #endif
         procfdtbl_clearfd(p, fd);
  
         while ((nfd = fdp->fd_lastfile) > 0 &&
  #endif
         procfdtbl_clearfd(p, fd);
  
         while ((nfd = fdp->fd_lastfile) > 0 &&
-                       fdp->fd_ofiles[nfd] == NULL &&
-                       !(fdp->fd_ofileflags[nfd] & UF_RESERVED))
+           fdp->fd_ofiles[nfd] == NULL &&
+           !(fdp->fd_ofileflags[nfd] & UF_RESERVED)) {
                 /* JMM - What about files with lingering EV_VANISHED knotes? */
                 fdp->fd_lastfile--;
                 /* JMM - What about files with lingering EV_VANISHED knotes? */
                 fdp->fd_lastfile--;
+       }
  }
  
  
  }
  
  
@@ -447,80 +632,85 @@ fd_rdwr(
         uint64_t base,
         int64_t len,
         enum uio_seg segflg,
         uint64_t base,
         int64_t len,
         enum uio_seg segflg,
-       off_t   offset,
-       int     io_flg,
+       off_t   offset,
+       int     io_flg,
         int64_t *aresid)
  {
         int64_t *aresid)
  {
-        struct fileproc *fp;
-       proc_t  p;
-        int error = 0;
+       struct fileproc *fp;
+       proc_t  p;
+       int error = 0;
         int flags = 0;
         int spacetype;
         uio_t auio = NULL;
         int flags = 0;
         int spacetype;
         uio_t auio = NULL;
-       char uio_buf[ UIO_SIZEOF(1) ];
+       char uio_buf[UIO_SIZEOF(1)];
         struct vfs_context context = *(vfs_context_current());
         struct vfs_context context = *(vfs_context_current());
-       bool wrote_some = false;
  
         p = current_proc();
  
  
         p = current_proc();
  
-        error = fp_lookup(p, fd, &fp, 0);
-        if (error)
-                return(error);
+       error = fp_lookup(p, fd, &fp, 0);
+       if (error) {
+               return error;
+       }
  
  
-       if (fp->f_type != DTYPE_VNODE && fp->f_type != DTYPE_PIPE && fp->f_type != DTYPE_SOCKET) {
+       switch (FILEGLOB_DTYPE(fp->fp_glob)) {
+       case DTYPE_VNODE:
+       case DTYPE_PIPE:
+       case DTYPE_SOCKET:
+               break;
+       default:
                 error = EINVAL;
                 goto out;
         }
         if (rw == UIO_WRITE && !(fp->f_flag & FWRITE)) {
                 error = EINVAL;
                 goto out;
         }
         if (rw == UIO_WRITE && !(fp->f_flag & FWRITE)) {
-                error = EBADF;
+               error = EBADF;
                 goto out;
         }
  
         if (rw == UIO_READ && !(fp->f_flag & FREAD)) {
                 goto out;
         }
  
         if (rw == UIO_READ && !(fp->f_flag & FREAD)) {
-               error = EBADF;
-               goto out;
+               error = EBADF;
+               goto out;
         }
  
         }
  
-       context.vc_ucred = fp->f_fglob->fg_cred;
+       context.vc_ucred = fp->fp_glob->fg_cred;
  
  
-       if (UIO_SEG_IS_USER_SPACE(segflg))
+       if (UIO_SEG_IS_USER_SPACE(segflg)) {
                 spacetype = proc_is64bit(p) ? UIO_USERSPACE64 : UIO_USERSPACE32;
                 spacetype = proc_is64bit(p) ? UIO_USERSPACE64 : UIO_USERSPACE32;
-       else
+       } else {
                 spacetype = UIO_SYSSPACE;
                 spacetype = UIO_SYSSPACE;
+       }
  
         auio = uio_createwithbuffer(1, offset, spacetype, rw, &uio_buf[0], sizeof(uio_buf));
  
  
         auio = uio_createwithbuffer(1, offset, spacetype, rw, &uio_buf[0], sizeof(uio_buf));
  
-       uio_addiov(auio, base, len);
+       uio_addiov(auio, (user_addr_t)base, (user_size_t)len);
  
  
-       if ( !(io_flg & IO_APPEND))
+       if (!(io_flg & IO_APPEND)) {
                 flags = FOF_OFFSET;
                 flags = FOF_OFFSET;
+       }
  
         if (rw == UIO_WRITE) {
                 user_ssize_t orig_resid = uio_resid(auio);
                 error = fo_write(fp, auio, flags, &context);
  
         if (rw == UIO_WRITE) {
                 user_ssize_t orig_resid = uio_resid(auio);
                 error = fo_write(fp, auio, flags, &context);
-               wrote_some = uio_resid(auio) < orig_resid;
-       } else
+               if (uio_resid(auio) < orig_resid) {
+                       os_atomic_or(&fp->fp_glob->fg_flag, FWASWRITTEN, relaxed);
+               }
+       } else {
                 error = fo_read(fp, auio, flags, &context);
                 error = fo_read(fp, auio, flags, &context);
+       }
  
  
-       if (aresid)
+       if (aresid) {
                 *aresid = uio_resid(auio);
                 *aresid = uio_resid(auio);
-       else {
-               if (uio_resid(auio) && error == 0)
-                       error = EIO;
+       } else if (uio_resid(auio) && error == 0) {
+               error = EIO;
         }
  out:
         }
  out:
-        if (wrote_some)
-                fp_drop_written(p, fd, fp);
-        else
-                fp_drop(p, fd, fp, 0);
-
+       fp_drop(p, fd, fp, 0);
         return error;
  }
  
  
  
  /*
         return error;
  }
  
  
  
  /*
- * dup
+ * sys_dup
   *
   * Description:        Duplicate a file descriptor.
   *
   *
   * Description:        Duplicate a file descriptor.
   *
@@ -535,7 +725,7 @@ out:
   *             *retval (modified)              The new descriptor
   */
  int
   *             *retval (modified)              The new descriptor
   */
  int
-dup(proc_t p, struct dup_args *uap, int32_t *retval)
+sys_dup(proc_t p, struct dup_args *uap, int32_t *retval)
  {
         struct filedesc *fdp = p->p_fd;
         int old = uap->fd;
  {
         struct filedesc *fdp = p->p_fd;
         int old = uap->fd;
@@ -543,35 +733,35 @@ dup(proc_t p, struct dup_args *uap, int32_t *retval)
         struct fileproc *fp;
  
         proc_fdlock(p);
         struct fileproc *fp;
  
         proc_fdlock(p);
-       if ( (error = fp_lookup(p, old, &fp, 1)) ) {
+       if ((error = fp_lookup(p, old, &fp, 1))) {
                 proc_fdunlock(p);
                 proc_fdunlock(p);
-               return(error);
+               return error;
         }
         }
-       if (FP_ISGUARDED(fp, GUARD_DUP)) {
+       if (fp_isguarded(fp, GUARD_DUP)) {
                 error = fp_guard_exception(p, old, fp, kGUARD_EXC_DUP);
                 (void) fp_drop(p, old, fp, 1);
                 proc_fdunlock(p);
                 error = fp_guard_exception(p, old, fp, kGUARD_EXC_DUP);
                 (void) fp_drop(p, old, fp, 1);
                 proc_fdunlock(p);
-               return (error);
+               return error;
         }
         }
-       if ( (error = fdalloc(p, 0, &new)) ) {
+       if ((error = fdalloc(p, 0, &new))) {
                 fp_drop(p, old, fp, 1);
                 proc_fdunlock(p);
                 fp_drop(p, old, fp, 1);
                 proc_fdunlock(p);
-               return (error);
+               return error;
         }
         error = finishdup(p, fdp, old, new, 0, retval);
         fp_drop(p, old, fp, 1);
         proc_fdunlock(p);
  
         }
         error = finishdup(p, fdp, old, new, 0, retval);
         fp_drop(p, old, fp, 1);
         proc_fdunlock(p);
  
-       if (ENTR_SHOULDTRACE && fp->f_type == DTYPE_SOCKET) {
+       if (ENTR_SHOULDTRACE && FILEGLOB_DTYPE(fp->fp_glob) == DTYPE_SOCKET) {
                 KERNEL_ENERGYTRACE(kEnTrActKernSocket, DBG_FUNC_START,
                     new, 0, (int64_t)VM_KERNEL_ADDRPERM(fp->f_data));
         }
  
                 KERNEL_ENERGYTRACE(kEnTrActKernSocket, DBG_FUNC_START,
                     new, 0, (int64_t)VM_KERNEL_ADDRPERM(fp->f_data));
         }
  
-       return (error);
+       return error;
  }
  
  /*
  }
  
  /*
- * dup2
+ * sys_dup2
   *
   * Description:        Duplicate a file descriptor to a particular value.
   *
   *
   * Description:        Duplicate a file descriptor to a particular value.
   *
@@ -587,44 +777,50 @@ dup(proc_t p, struct dup_args *uap, int32_t *retval)
   *             *retval (modified)              The new descriptor
   */
  int
   *             *retval (modified)              The new descriptor
   */
  int
-dup2(proc_t p, struct dup2_args *uap, int32_t *retval)
+sys_dup2(proc_t p, struct dup2_args *uap, int32_t *retval)
+{
+       return dup2(p, uap->from, uap->to, retval);
+}
+
+int
+dup2(proc_t p, int old, int new, int *retval)
  {
         struct filedesc *fdp = p->p_fd;
  {
         struct filedesc *fdp = p->p_fd;
-       int old = uap->from, new = uap->to;
-       int i, error;
         struct fileproc *fp, *nfp;
         struct fileproc *fp, *nfp;
+       int i, error;
+       rlim_t nofile = proc_limitgetcur(p, RLIMIT_NOFILE, TRUE);
  
         proc_fdlock(p);
  
  startover:
  
         proc_fdlock(p);
  
  startover:
-       if ( (error = fp_lookup(p, old, &fp, 1)) ) {
+       if ((error = fp_lookup(p, old, &fp, 1))) {
                 proc_fdunlock(p);
                 proc_fdunlock(p);
-               return(error);
+               return error;
         }
         }
-       if (FP_ISGUARDED(fp, GUARD_DUP)) {
+       if (fp_isguarded(fp, GUARD_DUP)) {
                 error = fp_guard_exception(p, old, fp, kGUARD_EXC_DUP);
                 (void) fp_drop(p, old, fp, 1);
                 proc_fdunlock(p);
                 error = fp_guard_exception(p, old, fp, kGUARD_EXC_DUP);
                 (void) fp_drop(p, old, fp, 1);
                 proc_fdunlock(p);
-               return (error);
+               return error;
         }
         if (new < 0 ||
         }
         if (new < 0 ||
-               (rlim_t)new >= p->p_rlimit[RLIMIT_NOFILE].rlim_cur ||
-           new >= maxfiles) {
+           (rlim_t)new >= nofile ||
+           new >= maxfilesperproc) {
                 fp_drop(p, old, fp, 1);
                 proc_fdunlock(p);
                 fp_drop(p, old, fp, 1);
                 proc_fdunlock(p);
-               return (EBADF);
+               return EBADF;
         }
         if (old == new) {
                 fp_drop(p, old, fp, 1);
                 *retval = new;
                 proc_fdunlock(p);
         }
         if (old == new) {
                 fp_drop(p, old, fp, 1);
                 *retval = new;
                 proc_fdunlock(p);
-               return (0);
+               return 0;
         }
         if (new < 0 || new >= fdp->fd_nfiles) {
         }
         if (new < 0 || new >= fdp->fd_nfiles) {
-               if ( (error = fdalloc(p, new, &i)) ) {
+               if ((error = fdalloc(p, new, &i))) {
                         fp_drop(p, old, fp, 1);
                         proc_fdunlock(p);
                         fp_drop(p, old, fp, 1);
                         proc_fdunlock(p);
-                       return (error);
+                       return error;
                 }
                 if (new != i) {
                         fdrelse(p, i);
                 }
                 if (new != i) {
                         fdrelse(p, i);
@@ -632,55 +828,48 @@ startover:
                 }
         } else {
  closeit:
                 }
         } else {
  closeit:
-               while ((fdp->fd_ofileflags[new] & UF_RESERVED) == UF_RESERVED)  {
-                               fp_drop(p, old, fp, 1);
-                               procfdtbl_waitfd(p, new);
+               if ((fdp->fd_ofileflags[new] & UF_RESERVED) == UF_RESERVED) {
+                       fp_drop(p, old, fp, 1);
+                       procfdtbl_waitfd(p, new);
  #if DIAGNOSTIC
  #if DIAGNOSTIC
-                               proc_fdlock_assert(p, LCK_MTX_ASSERT_OWNED);
+                       proc_fdlock_assert(p, LCK_MTX_ASSERT_OWNED);
  #endif
  #endif
-                               goto startover;
+                       goto startover;
                 }
  
                 }
  
-               if ((fdp->fd_ofiles[new] != NULL) &&
-                   ((error = fp_lookup(p, new, &nfp, 1)) == 0)) {
-                       fp_drop(p, old, fp, 1);
-                       if (FP_ISGUARDED(nfp, GUARD_CLOSE)) {
+               if ((nfp = fdp->fd_ofiles[new]) != NULL) {
+                       if (fp_isguarded(nfp, GUARD_CLOSE)) {
+                               fp_drop(p, old, fp, 1);
                                 error = fp_guard_exception(p,
                                     new, nfp, kGUARD_EXC_CLOSE);
                                 error = fp_guard_exception(p,
                                     new, nfp, kGUARD_EXC_CLOSE);
-                               (void) fp_drop(p, new, nfp, 1);
                                 proc_fdunlock(p);
                                 proc_fdunlock(p);
-                               return (error);
+                               return error;
                         }
                         }
-                       (void)close_internal_locked(p, new, nfp, FD_DUP2RESV);
-#if DIAGNOSTIC
-                       proc_fdlock_assert(p, LCK_MTX_ASSERT_OWNED);
-#endif
-                       procfdtbl_clearfd(p, new);
-                       goto startover;
-               } else  {
+                       (void)fp_close_and_unlock(p, new, nfp, FD_DUP2RESV);
+                       proc_fdlock(p);
+                       assert(fdp->fd_ofileflags[new] & UF_RESERVED);
+               } else {
  #if DIAGNOSTIC
  #if DIAGNOSTIC
-                       if (fdp->fd_ofiles[new] != NULL)
+                       if (fdp->fd_ofiles[new] != NULL) {
                                 panic("dup2: no ref on fileproc %d", new);
                                 panic("dup2: no ref on fileproc %d", new);
+                       }
  #endif
                         procfdtbl_reservefd(p, new);
                 }
  #endif
                         procfdtbl_reservefd(p, new);
                 }
-
-#if DIAGNOSTIC
-               proc_fdlock_assert(p, LCK_MTX_ASSERT_OWNED);
-#endif
-
         }
  #if DIAGNOSTIC
         }
  #if DIAGNOSTIC
-       if (fdp->fd_ofiles[new] != 0)
+       if (fdp->fd_ofiles[new] != 0) {
                 panic("dup2: overwriting fd_ofiles with new %d", new);
                 panic("dup2: overwriting fd_ofiles with new %d", new);
-       if ((fdp->fd_ofileflags[new] & UF_RESERVED) == 0)
+       }
+       if ((fdp->fd_ofileflags[new] & UF_RESERVED) == 0) {
                 panic("dup2: unreserved fileflags with new %d", new);
                 panic("dup2: unreserved fileflags with new %d", new);
+       }
  #endif
         error = finishdup(p, fdp, old, new, 0, retval);
         fp_drop(p, old, fp, 1);
         proc_fdunlock(p);
  
  #endif
         error = finishdup(p, fdp, old, new, 0, retval);
         fp_drop(p, old, fp, 1);
         proc_fdunlock(p);
  
-       return(error);
+       return error;
  }
  
  
  }
  
  
@@ -706,15 +895,17 @@ closeit:
   *             blocking operation.
   */
  int
   *             blocking operation.
   */
  int
-fcntl(proc_t p, struct fcntl_args *uap, int32_t *retval)
+sys_fcntl(proc_t p, struct fcntl_args *uap, int32_t *retval)
  {
         __pthread_testcancel(1);
  {
         __pthread_testcancel(1);
-       return(fcntl_nocancel(p, (struct fcntl_nocancel_args *)uap, retval));
+       return sys_fcntl_nocancel(p, (struct fcntl_nocancel_args *)uap, retval);
  }
  
  }
  
+#define ACCOUNT_OPENFROM_ENTITLEMENT \
+       "com.apple.private.vfs.role-account-openfrom"
  
  /*
  
  /*
- * fcntl_nocancel
+ * sys_fcntl_nocancel
   *
   * Description:        A non-cancel-testing file control system call.
   *
   *
   * Description:        A non-cancel-testing file control system call.
   *
@@ -757,6 +948,7 @@ fcntl(proc_t p, struct fcntl_args *uap, int32_t *retval)
   *     VNOP_ALLOCATE:???
   * [F_SETSIZE,F_RDADVISE]
   *             EBADF
   *     VNOP_ALLOCATE:???
   * [F_SETSIZE,F_RDADVISE]
   *             EBADF
+ *             EINVAL
   *     copyin:EFAULT
   *     vnode_getwithref:???
   * [F_RDAHEAD,F_NOCACHE]
   *     copyin:EFAULT
   *     vnode_getwithref:???
   * [F_RDAHEAD,F_NOCACHE]
@@ -768,13 +960,14 @@ fcntl(proc_t p, struct fcntl_args *uap, int32_t *retval)
   *             *retval (modified)              fcntl return value (if any)
   */
  int
   *             *retval (modified)              fcntl return value (if any)
   */
  int
-fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
+sys_fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  {
         int fd = uap->fd;
         struct filedesc *fdp = p->p_fd;
         struct fileproc *fp;
         char *pop;
  {
         int fd = uap->fd;
         struct filedesc *fdp = p->p_fd;
         struct fileproc *fp;
         char *pop;
-       struct vnode *vp = NULLVP;      /* for AUDIT_ARG() at end */
+       struct vnode *vp = NULLVP;      /* for AUDIT_ARG() at end */
+       unsigned int oflags, nflags;
         int i, tmp, error, error2, flg = 0;
         struct flock fl = {};
         struct flocktimeout fltimeout;
         int i, tmp, error, error2, flg = 0;
         struct flock fl = {};
         struct flocktimeout fltimeout;
@@ -786,14 +979,18 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
         unsigned int fflag;
         user_addr_t argp;
         boolean_t is64bit;
         unsigned int fflag;
         user_addr_t argp;
         boolean_t is64bit;
+       rlim_t nofile;
+       int has_entitlement = 0;
  
         AUDIT_ARG(fd, uap->fd);
         AUDIT_ARG(cmd, uap->cmd);
  
  
         AUDIT_ARG(fd, uap->fd);
         AUDIT_ARG(cmd, uap->cmd);
  
+       nofile = proc_limitgetcur(p, RLIMIT_NOFILE, TRUE);
+
         proc_fdlock(p);
         proc_fdlock(p);
-       if ( (error = fp_lookup(p, fd, &fp, 1)) ) {
+       if ((error = fp_lookup(p, fd, &fp, 1))) {
                 proc_fdunlock(p);
                 proc_fdunlock(p);
-               return(error);
+               return error;
         }
         context.vc_thread = current_thread();
         context.vc_ucred = fp->f_cred;
         }
         context.vc_thread = current_thread();
         context.vc_ucred = fp->f_cred;
@@ -801,8 +998,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
         is64bit = proc_is64bit(p);
         if (is64bit) {
                 argp = uap->arg;
         is64bit = proc_is64bit(p);
         if (is64bit) {
                 argp = uap->arg;
-       }
-       else {
+       } else {
                 /*
                  * Since the arg parameter is defined as a long but may be
                  * either a long or a pointer we must take care to handle
                 /*
                  * Since the arg parameter is defined as a long but may be
                  * either a long or a pointer we must take care to handle
@@ -815,32 +1011,33 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 argp = CAST_USER_ADDR_T((uint32_t)uap->arg);
         }
  
                 argp = CAST_USER_ADDR_T((uint32_t)uap->arg);
         }
  
-       pop = &fdp->fd_ofileflags[fd];
-
  #if CONFIG_MACF
  #if CONFIG_MACF
-       error = mac_file_check_fcntl(proc_ucred(p), fp->f_fglob, uap->cmd,
+       error = mac_file_check_fcntl(proc_ucred(p), fp->fp_glob, uap->cmd,
             uap->arg);
             uap->arg);
-       if (error)
+       if (error) {
                 goto out;
                 goto out;
+       }
  #endif
  
  #endif
  
-       switch (uap->cmd) {
+       pop = &fdp->fd_ofileflags[fd];
  
  
+       switch (uap->cmd) {
         case F_DUPFD:
         case F_DUPFD_CLOEXEC:
         case F_DUPFD:
         case F_DUPFD_CLOEXEC:
-               if (FP_ISGUARDED(fp, GUARD_DUP)) {
+               if (fp_isguarded(fp, GUARD_DUP)) {
                         error = fp_guard_exception(p, fd, fp, kGUARD_EXC_DUP);
                         goto out;
                 }
                 newmin = CAST_DOWN_EXPLICIT(int, uap->arg); /* arg is an int, so we won't lose bits */
                 AUDIT_ARG(value32, newmin);
                         error = fp_guard_exception(p, fd, fp, kGUARD_EXC_DUP);
                         goto out;
                 }
                 newmin = CAST_DOWN_EXPLICIT(int, uap->arg); /* arg is an int, so we won't lose bits */
                 AUDIT_ARG(value32, newmin);
-               if ((u_int)newmin >= p->p_rlimit[RLIMIT_NOFILE].rlim_cur ||
-                   newmin >= maxfiles) {
+               if ((rlim_t)newmin >= nofile ||
+                   newmin >= maxfilesperproc) {
                         error = EINVAL;
                         goto out;
                 }
                         error = EINVAL;
                         goto out;
                 }
-               if ( (error = fdalloc(p, newmin, &i)) )
+               if ((error = fdalloc(p, newmin, &i))) {
                         goto out;
                         goto out;
+               }
                 error = finishdup(p, fdp, fd, i,
                     uap->cmd == F_DUPFD_CLOEXEC ? UF_EXCLOSE : 0, retval);
                 goto out;
                 error = finishdup(p, fdp, fd, i,
                     uap->cmd == F_DUPFD_CLOEXEC ? UF_EXCLOSE : 0, retval);
                 goto out;
@@ -851,11 +1048,11 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 goto out;
  
         case F_SETFD:
                 goto out;
  
         case F_SETFD:
-               AUDIT_ARG(value32, uap->arg);
-               if (uap->arg & FD_CLOEXEC)
+               AUDIT_ARG(value32, (uint32_t)uap->arg);
+               if (uap->arg & FD_CLOEXEC) {
                         *pop |= UF_EXCLOSE;
                         *pop |= UF_EXCLOSE;
-               else {
-                       if (FILEPROC_TYPE(fp) == FTYPE_GUARDED) {
+               } else {
+                       if (fp_isguarded(fp, 0)) {
                                 error = fp_guard_exception(p,
                                     fd, fp, kGUARD_EXC_NOCLOEXEC);
                                 goto out;
                                 error = fp_guard_exception(p,
                                     fd, fp, kGUARD_EXC_NOCLOEXEC);
                                 goto out;
@@ -871,19 +1068,29 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 goto out;
  
         case F_SETFL:
                 goto out;
  
         case F_SETFL:
-               fp->f_flag &= ~FCNTLFLAGS;
+               // FIXME (rdar://54898652)
+               //
+               // this code is broken if fnctl(F_SETFL), ioctl() are
+               // called concurrently for the same fileglob.
+
                 tmp = CAST_DOWN_EXPLICIT(int, uap->arg); /* arg is an int, so we won't lose bits */
                 AUDIT_ARG(value32, tmp);
                 tmp = CAST_DOWN_EXPLICIT(int, uap->arg); /* arg is an int, so we won't lose bits */
                 AUDIT_ARG(value32, tmp);
-               fp->f_flag |= FFLAGS(tmp) & FCNTLFLAGS;
-               tmp = fp->f_flag & FNONBLOCK;
+
+               os_atomic_rmw_loop(&fp->f_flag, oflags, nflags, relaxed, {
+                       nflags  = oflags & ~FCNTLFLAGS;
+                       nflags |= FFLAGS(tmp) & FCNTLFLAGS;
+               });
+               tmp = nflags & FNONBLOCK;
                 error = fo_ioctl(fp, FIONBIO, (caddr_t)&tmp, &context);
                 error = fo_ioctl(fp, FIONBIO, (caddr_t)&tmp, &context);
-               if (error)
+               if (error) {
                         goto out;
                         goto out;
-               tmp = fp->f_flag & FASYNC;
+               }
+               tmp = nflags & FASYNC;
                 error = fo_ioctl(fp, FIOASYNC, (caddr_t)&tmp, &context);
                 error = fo_ioctl(fp, FIOASYNC, (caddr_t)&tmp, &context);
-               if (!error)
+               if (!error) {
                         goto out;
                         goto out;
-               fp->f_flag &= ~FNONBLOCK;
+               }
+               os_atomic_andnot(&fp->f_flag, FNONBLOCK, relaxed);
                 tmp = 0;
                 (void)fo_ioctl(fp, FIONBIO, (caddr_t)&tmp, &context);
                 goto out;
                 tmp = 0;
                 (void)fo_ioctl(fp, FIONBIO, (caddr_t)&tmp, &context);
                 goto out;
@@ -894,7 +1101,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         error = 0;
                         goto out;
                 }
                         error = 0;
                         goto out;
                 }
-               error = fo_ioctl(fp, (int)TIOCGPGRP, (caddr_t)retval, &context);
+               error = fo_ioctl(fp, TIOCGPGRP, (caddr_t)retval, &context);
                 *retval = -*retval;
                 goto out;
  
                 *retval = -*retval;
                 goto out;
  
@@ -903,7 +1110,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 AUDIT_ARG(value32, tmp);
                 if (fp->f_type == DTYPE_SOCKET) {
                         ((struct socket *)fp->f_data)->so_pgid = tmp;
                 AUDIT_ARG(value32, tmp);
                 if (fp->f_type == DTYPE_SOCKET) {
                         ((struct socket *)fp->f_data)->so_pgid = tmp;
-                       error =0;
+                       error = 0;
                         goto out;
                 }
                 if (fp->f_type == DTYPE_PIPE) {
                         goto out;
                 }
                 if (fp->f_type == DTYPE_PIPE) {
@@ -922,7 +1129,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         tmp = (int)p1->p_pgrpid;
                         proc_rele(p1);
                 }
                         tmp = (int)p1->p_pgrpid;
                         proc_rele(p1);
                 }
-               error =  fo_ioctl(fp, (int)TIOCSPGRP, (caddr_t)&tmp, &context);
+               error =  fo_ioctl(fp, TIOCSPGRP, (caddr_t)&tmp, &context);
                 goto out;
  
         case F_SETNOSIGPIPE:
                 goto out;
  
         case F_SETNOSIGPIPE:
@@ -930,18 +1137,19 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 if (fp->f_type == DTYPE_SOCKET) {
  #if SOCKETS
                         error = sock_setsockopt((struct socket *)fp->f_data,
                 if (fp->f_type == DTYPE_SOCKET) {
  #if SOCKETS
                         error = sock_setsockopt((struct socket *)fp->f_data,
-                           SOL_SOCKET, SO_NOSIGPIPE, &tmp, sizeof (tmp));
+                           SOL_SOCKET, SO_NOSIGPIPE, &tmp, sizeof(tmp));
  #else
                         error = EINVAL;
  #endif
                 } else {
  #else
                         error = EINVAL;
  #endif
                 } else {
-                       struct fileglob *fg = fp->f_fglob;
+                       struct fileglob *fg = fp->fp_glob;
  
                         lck_mtx_lock_spin(&fg->fg_lock);
  
                         lck_mtx_lock_spin(&fg->fg_lock);
-                       if (tmp)
+                       if (tmp) {
                                 fg->fg_lflags |= FG_NOSIGPIPE;
                                 fg->fg_lflags |= FG_NOSIGPIPE;
-                       else
-                               fg->fg_lflags &= FG_NOSIGPIPE;
+                       } else {
+                               fg->fg_lflags &= ~FG_NOSIGPIPE;
+                       }
                         lck_mtx_unlock(&fg->fg_lock);
                         error = 0;
                 }
                         lck_mtx_unlock(&fg->fg_lock);
                         error = 0;
                 }
@@ -950,15 +1158,15 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
         case F_GETNOSIGPIPE:
                 if (fp->f_type == DTYPE_SOCKET) {
  #if SOCKETS
         case F_GETNOSIGPIPE:
                 if (fp->f_type == DTYPE_SOCKET) {
  #if SOCKETS
-                       int retsize = sizeof (*retval);
+                       int retsize = sizeof(*retval);
                         error = sock_getsockopt((struct socket *)fp->f_data,
                             SOL_SOCKET, SO_NOSIGPIPE, retval, &retsize);
  #else
                         error = EINVAL;
  #endif
                 } else {
                         error = sock_getsockopt((struct socket *)fp->f_data,
                             SOL_SOCKET, SO_NOSIGPIPE, retval, &retsize);
  #else
                         error = EINVAL;
  #endif
                 } else {
-                       *retval = (fp->f_fglob->fg_lflags & FG_NOSIGPIPE) ?
-                               1 : 0;
+                       *retval = (fp->fp_glob->fg_lflags & FG_NOSIGPIPE) ?
+                           1 : 0;
                         error = 0;
                 }
                 goto out;
                         error = 0;
                 }
                 goto out;
@@ -972,30 +1180,31 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                  * mechanism to move the descriptor elsewhere will fail.
                  */
                 if (CAST_DOWN_EXPLICIT(int, uap->arg)) {
                  * mechanism to move the descriptor elsewhere will fail.
                  */
                 if (CAST_DOWN_EXPLICIT(int, uap->arg)) {
-                       struct fileglob *fg = fp->f_fglob;
+                       struct fileglob *fg = fp->fp_glob;
  
                         lck_mtx_lock_spin(&fg->fg_lock);
  
                         lck_mtx_lock_spin(&fg->fg_lock);
-                       if (fg->fg_lflags & FG_CONFINED)
+                       if (fg->fg_lflags & FG_CONFINED) {
                                 error = 0;
                                 error = 0;
-                       else if (1 != fg->fg_count)
-                               error = EAGAIN; /* go close the dup .. */
-                       else if (UF_FORKCLOSE == (*pop & UF_FORKCLOSE)) {
+                       } else if (1 != os_ref_get_count_raw(&fg->fg_count)) {
+                               error = EAGAIN; /* go close the dup .. */
+                       } else if (UF_FORKCLOSE == (*pop & UF_FORKCLOSE)) {
                                 fg->fg_lflags |= FG_CONFINED;
                                 error = 0;
                                 fg->fg_lflags |= FG_CONFINED;
                                 error = 0;
-                       } else
-                               error = EBADF;  /* open without O_CLOFORK? */
+                       } else {
+                               error = EBADF;  /* open without O_CLOFORK? */
+                       }
                         lck_mtx_unlock(&fg->fg_lock);
                 } else {
                         /*
                          * Other subsystems may have built on the immutability
                          * of FG_CONFINED; clearing it may be tricky.
                          */
                         lck_mtx_unlock(&fg->fg_lock);
                 } else {
                         /*
                          * Other subsystems may have built on the immutability
                          * of FG_CONFINED; clearing it may be tricky.
                          */
-                       error = EPERM;          /* immutable */
+                       error = EPERM;          /* immutable */
                 }
                 goto out;
  
         case F_GETCONFINED:
                 }
                 goto out;
  
         case F_GETCONFINED:
-               *retval = (fp->f_fglob->fg_lflags & FG_CONFINED) ? 1 : 0;
+               *retval = (fp->fp_glob->fg_lflags & FG_CONFINED) ? 1 : 0;
                 error = 0;
                 goto out;
  
                 error = 0;
                 goto out;
  
@@ -1004,7 +1213,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
         case F_OFD_SETLKWTIMEOUT:
         case F_OFD_SETLKW:
                 flg |= F_WAIT;
         case F_OFD_SETLKWTIMEOUT:
         case F_OFD_SETLKW:
                 flg |= F_WAIT;
-               /* Fall into F_SETLK */
+               OS_FALLTHROUGH;
  
         case F_SETLK:
         case F_OFD_SETLK:
  
         case F_SETLK:
         case F_OFD_SETLK:
@@ -1041,14 +1250,15 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         goto outdrop;
                 }
  
                         goto outdrop;
                 }
  
-               if ( (error = vnode_getwithref(vp)) ) {
+               if ((error = vnode_getwithref(vp))) {
                         goto outdrop;
                 }
                         goto outdrop;
                 }
-               if (fl.l_whence == SEEK_CUR)
+               if (fl.l_whence == SEEK_CUR) {
                         fl.l_start += offset;
                         fl.l_start += offset;
+               }
  
  #if CONFIG_MACF
  
  #if CONFIG_MACF
-               error = mac_file_check_lock(proc_ucred(p), fp->f_fglob,
+               error = mac_file_check_lock(proc_ucred(p), fp->fp_glob,
                     F_SETLK, &fl);
                 if (error) {
                         (void)vnode_put(vp);
                     F_SETLK, &fl);
                 if (error) {
                         (void)vnode_put(vp);
@@ -1066,7 +1276,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                                         error = EBADF;
                                         break;
                                 }
                                         error = EBADF;
                                         break;
                                 }
-                               error = VNOP_ADVLOCK(vp, (caddr_t)fp->f_fglob,
+                               error = VNOP_ADVLOCK(vp, (caddr_t)fp->fp_glob,
                                     F_SETLK, &fl, flg, &context, timeout);
                                 break;
                         case F_WRLCK:
                                     F_SETLK, &fl, flg, &context, timeout);
                                 break;
                         case F_WRLCK:
@@ -1074,11 +1284,11 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                                         error = EBADF;
                                         break;
                                 }
                                         error = EBADF;
                                         break;
                                 }
-                               error = VNOP_ADVLOCK(vp, (caddr_t)fp->f_fglob,
+                               error = VNOP_ADVLOCK(vp, (caddr_t)fp->fp_glob,
                                     F_SETLK, &fl, flg, &context, timeout);
                                 break;
                         case F_UNLCK:
                                     F_SETLK, &fl, flg, &context, timeout);
                                 break;
                         case F_UNLCK:
-                               error = VNOP_ADVLOCK(vp, (caddr_t)fp->f_fglob,
+                               error = VNOP_ADVLOCK(vp, (caddr_t)fp->fp_glob,
                                     F_UNLCK, &fl, F_OFD_LOCK, &context,
                                     timeout);
                                 break;
                                     F_UNLCK, &fl, F_OFD_LOCK, &context,
                                     timeout);
                                 break;
@@ -1088,7 +1298,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         }
                         if (0 == error &&
                             (F_RDLCK == fl.l_type || F_WRLCK == fl.l_type)) {
                         }
                         if (0 == error &&
                             (F_RDLCK == fl.l_type || F_WRLCK == fl.l_type)) {
-                               struct fileglob *fg = fp->f_fglob;
+                               struct fileglob *fg = fp->fp_glob;
  
                                 /*
                                  * arrange F_UNLCK on last close (once
  
                                 /*
                                  * arrange F_UNLCK on last close (once
@@ -1110,7 +1320,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                                         break;
                                 }
                                 // XXX UInt32 unsafe for LP64 kernel
                                         break;
                                 }
                                 // XXX UInt32 unsafe for LP64 kernel
-                               OSBitOrAtomic(P_LADVLOCK, &p->p_ladvflag);
+                               os_atomic_or(&p->p_ladvflag, P_LADVLOCK, relaxed);
                                 error = VNOP_ADVLOCK(vp, (caddr_t)p,
                                     F_SETLK, &fl, flg, &context, timeout);
                                 break;
                                 error = VNOP_ADVLOCK(vp, (caddr_t)p,
                                     F_SETLK, &fl, flg, &context, timeout);
                                 break;
@@ -1120,7 +1330,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                                         break;
                                 }
                                 // XXX UInt32 unsafe for LP64 kernel
                                         break;
                                 }
                                 // XXX UInt32 unsafe for LP64 kernel
-                               OSBitOrAtomic(P_LADVLOCK, &p->p_ladvflag);
+                               os_atomic_or(&p->p_ladvflag, P_LADVLOCK, relaxed);
                                 error = VNOP_ADVLOCK(vp, (caddr_t)p,
                                     F_SETLK, &fl, flg, &context, timeout);
                                 break;
                                 error = VNOP_ADVLOCK(vp, (caddr_t)p,
                                     F_SETLK, &fl, flg, &context, timeout);
                                 break;
@@ -1152,8 +1362,9 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  
                 /* Copy in the lock structure */
                 error = copyin(argp, (caddr_t)&fl, sizeof(fl));
  
                 /* Copy in the lock structure */
                 error = copyin(argp, (caddr_t)&fl, sizeof(fl));
-               if (error)
+               if (error) {
                         goto outdrop;
                         goto outdrop;
+               }
  
                 /* Check starting byte and ending byte for EOVERFLOW in SEEK_CUR */
                 /* and ending byte for EOVERFLOW in SEEK_SET */
  
                 /* Check starting byte and ending byte for EOVERFLOW in SEEK_CUR */
                 /* and ending byte for EOVERFLOW in SEEK_SET */
@@ -1187,22 +1398,23 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         goto outdrop;
                 }
  
                         goto outdrop;
                 }
  
-               if ( (error = vnode_getwithref(vp)) == 0 ) {
-                       if (fl.l_whence == SEEK_CUR)
-                               fl.l_start += offset;
+               if ((error = vnode_getwithref(vp)) == 0) {
+                       if (fl.l_whence == SEEK_CUR) {
+                               fl.l_start += offset;
+                       }
  
  #if CONFIG_MACF
  
  #if CONFIG_MACF
-                       error = mac_file_check_lock(proc_ucred(p), fp->f_fglob,
+                       error = mac_file_check_lock(proc_ucred(p), fp->fp_glob,
                             uap->cmd, &fl);
                         if (error == 0)
  #endif
                         switch (uap->cmd) {
                         case F_OFD_GETLK:
                             uap->cmd, &fl);
                         if (error == 0)
  #endif
                         switch (uap->cmd) {
                         case F_OFD_GETLK:
-                               error = VNOP_ADVLOCK(vp, (caddr_t)fp->f_fglob,
+                               error = VNOP_ADVLOCK(vp, (caddr_t)fp->fp_glob,
                                     F_GETLK, &fl, F_OFD_LOCK, &context, NULL);
                                 break;
                         case F_OFD_GETLKPID:
                                     F_GETLK, &fl, F_OFD_LOCK, &context, NULL);
                                 break;
                         case F_OFD_GETLKPID:
-                               error = VNOP_ADVLOCK(vp, (caddr_t)fp->f_fglob,
+                               error = VNOP_ADVLOCK(vp, (caddr_t)fp->fp_glob,
                                     F_GETLKPID, &fl, F_OFD_LOCK, &context, NULL);
                                 break;
                         default:
                                     F_GETLKPID, &fl, F_OFD_LOCK, &context, NULL);
                                 break;
                         default:
@@ -1213,8 +1425,9 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  
                         (void)vnode_put(vp);
  
  
                         (void)vnode_put(vp);
  
-                       if (error == 0)
+                       if (error == 0) {
                                 error = copyout((caddr_t)&fl, argp, sizeof(fl));
                                 error = copyout((caddr_t)&fl, argp, sizeof(fl));
+                       }
                 }
                 goto outdrop;
  
                 }
                 goto outdrop;
  
@@ -1237,8 +1450,9 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 }
  
                 error = copyin(argp, (caddr_t)&alloc_struct, sizeof(alloc_struct));
                 }
  
                 error = copyin(argp, (caddr_t)&alloc_struct, sizeof(alloc_struct));
-               if (error)
+               if (error) {
                         goto outdrop;
                         goto outdrop;
+               }
  
                 /* now set the space allocated to 0 */
                 alloc_struct.fst_bytesalloc = 0;
  
                 /* now set the space allocated to 0 */
                 alloc_struct.fst_bytesalloc = 0;
@@ -1251,11 +1465,13 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  
                 alloc_flags |= PREALLOCATE;
  
  
                 alloc_flags |= PREALLOCATE;
  
-               if (alloc_struct.fst_flags & F_ALLOCATECONTIG)
+               if (alloc_struct.fst_flags & F_ALLOCATECONTIG) {
                         alloc_flags |= ALLOCATECONTIG;
                         alloc_flags |= ALLOCATECONTIG;
+               }
  
  
-               if (alloc_struct.fst_flags & F_ALLOCATEALL)
+               if (alloc_struct.fst_flags & F_ALLOCATEALL) {
                         alloc_flags |= ALLOCATEALL;
                         alloc_flags |= ALLOCATEALL;
+               }
  
                 /*
                  * Do any position mode specific stuff.  The only
  
                 /*
                  * Do any position mode specific stuff.  The only
@@ -1263,7 +1479,6 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                  */
  
                 switch (alloc_struct.fst_posmode) {
                  */
  
                 switch (alloc_struct.fst_posmode) {
-
                 case F_PEOFPOSMODE:
                         if (alloc_struct.fst_offset != 0) {
                                 error = EINVAL;
                 case F_PEOFPOSMODE:
                         if (alloc_struct.fst_offset != 0) {
                                 error = EINVAL;
@@ -1285,21 +1500,22 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 default: {
                         error = EINVAL;
                         goto outdrop;
                 default: {
                         error = EINVAL;
                         goto outdrop;
-                       }
                 }
                 }
-               if ( (error = vnode_getwithref(vp)) == 0 ) {
-                       /*
+               }
+               if ((error = vnode_getwithref(vp)) == 0) {
+                       /*
                          * call allocate to get the space
                          */
                          * call allocate to get the space
                          */
-                       error = VNOP_ALLOCATE(vp,alloc_struct.fst_length,alloc_flags,
-                                             &alloc_struct.fst_bytesalloc, alloc_struct.fst_offset,
-                                             &context);
+                       error = VNOP_ALLOCATE(vp, alloc_struct.fst_length, alloc_flags,
+                           &alloc_struct.fst_bytesalloc, alloc_struct.fst_offset,
+                           &context);
                         (void)vnode_put(vp);
  
                         error2 = copyout((caddr_t)&alloc_struct, argp, sizeof(alloc_struct));
  
                         (void)vnode_put(vp);
  
                         error2 = copyout((caddr_t)&alloc_struct, argp, sizeof(alloc_struct));
  
-                       if (error == 0)
+                       if (error == 0) {
                                 error = error2;
                                 error = error2;
+                       }
                 }
                 goto outdrop;
         }
                 }
                 goto outdrop;
         }
@@ -1329,7 +1545,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 }
  
  #if CONFIG_MACF
                 }
  
  #if CONFIG_MACF
-               if ((error = mac_vnode_check_write(&context, fp->f_fglob->fg_cred, vp))) {
+               if ((error = mac_vnode_check_write(&context, fp->fp_glob->fg_cred, vp))) {
                         (void)vnode_put(vp);
                         goto outdrop;
                 }
                         (void)vnode_put(vp);
                         goto outdrop;
                 }
@@ -1375,6 +1591,50 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  
                 goto outdrop;
         }
  
                 goto outdrop;
         }
+       case F_SPECULATIVE_READ: {
+               fspecread_t args;
+
+               if (fp->f_type != DTYPE_VNODE) {
+                       error = EBADF;
+                       goto out;
+               }
+
+               vp = (struct vnode *)fp->f_data;
+               proc_fdunlock(p);
+
+               if ((error = copyin(argp, (caddr_t)&args, sizeof(args)))) {
+                       goto outdrop;
+               }
+
+               /* Discard invalid offsets or lengths */
+               if ((args.fsr_offset < 0) || (args.fsr_length < 0)) {
+                       error = EINVAL;
+                       goto outdrop;
+               }
+
+               /*
+                * Round the file offset down to a page-size boundary (or to 0).
+                * The filesystem will need to round the length up to the end of the page boundary
+                * or to the EOF of the file.
+                */
+               uint64_t foff = (((uint64_t)args.fsr_offset) & ~((uint64_t)PAGE_MASK));
+               uint64_t foff_delta = args.fsr_offset - foff;
+               args.fsr_offset = (off_t) foff;
+
+               /*
+                * Now add in the delta to the supplied length. Since we may have adjusted the
+                * offset, increase it by the amount that we adjusted.
+                */
+               args.fsr_length += foff_delta;
+
+               if ((error = vnode_getwithref(vp))) {
+                       goto outdrop;
+               }
+               error = VNOP_IOCTL(vp, F_SPECULATIVE_READ, (caddr_t)&args, 0, &context);
+               (void)vnode_put(vp);
+
+               goto outdrop;
+       }
         case F_SETSIZE:
                 if (fp->f_type != DTYPE_VNODE) {
                         error = EBADF;
         case F_SETSIZE:
                 if (fp->f_type != DTYPE_VNODE) {
                         error = EBADF;
@@ -1383,18 +1643,20 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 vp = (struct vnode *)fp->f_data;
                 proc_fdunlock(p);
  
                 vp = (struct vnode *)fp->f_data;
                 proc_fdunlock(p);
  
-               error = copyin(argp, (caddr_t)&offset, sizeof (off_t));
-               if (error)
+               error = copyin(argp, (caddr_t)&offset, sizeof(off_t));
+               if (error) {
                         goto outdrop;
                         goto outdrop;
+               }
                 AUDIT_ARG(value64, offset);
  
                 error = vnode_getwithref(vp);
                 AUDIT_ARG(value64, offset);
  
                 error = vnode_getwithref(vp);
-               if (error)
+               if (error) {
                         goto outdrop;
                         goto outdrop;
+               }
  
  #if CONFIG_MACF
                 error = mac_vnode_check_truncate(&context,
  
  #if CONFIG_MACF
                 error = mac_vnode_check_truncate(&context,
-                   fp->f_fglob->fg_cred, vp);
+                   fp->fp_glob->fg_cred, vp);
                 if (error) {
                         (void)vnode_put(vp);
                         goto outdrop;
                 if (error) {
                         (void)vnode_put(vp);
                         goto outdrop;
@@ -1402,21 +1664,26 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  #endif
                 /*
                  * Make sure that we are root.  Growing a file
  #endif
                 /*
                  * Make sure that we are root.  Growing a file
-                * without zero filling the data is a security hole
-                * root would have access anyway so we'll allow it
+                * without zero filling the data is a security hole.
                  */
                 if (!kauth_cred_issuser(kauth_cred_get())) {
                         error = EACCES;
                 } else {
                         /*
                  */
                 if (!kauth_cred_issuser(kauth_cred_get())) {
                         error = EACCES;
                 } else {
                         /*
-                        * set the file size
+                        * Require privilege to change file size without zerofill,
+                        * else will change the file size and zerofill it.
                          */
                          */
-                       error = vnode_setsize(vp, offset, IO_NOZEROFILL,
-                           &context);
+                       error = priv_check_cred(kauth_cred_get(), PRIV_VFS_SETSIZE, 0);
+                       if (error == 0) {
+                               error = vnode_setsize(vp, offset, IO_NOZEROFILL, &context);
+                       } else {
+                               error = vnode_setsize(vp, offset, 0, &context);
+                       }
  
  #if CONFIG_MACF
  
  #if CONFIG_MACF
-                       if (error == 0)
-                               mac_vnode_notify_truncate(&context, fp->f_fglob->fg_cred, vp);
+                       if (error == 0) {
+                               mac_vnode_notify_truncate(&context, fp->fp_glob->fg_cred, vp);
+                       }
  #endif
                 }
  
  #endif
                 }
  
@@ -1428,11 +1695,11 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         error = EBADF;
                         goto out;
                 }
                         error = EBADF;
                         goto out;
                 }
-               if (uap->arg)
-                       fp->f_fglob->fg_flag &= ~FNORDAHEAD;
-               else
-                       fp->f_fglob->fg_flag |= FNORDAHEAD;
-
+               if (uap->arg) {
+                       os_atomic_andnot(&fp->fp_glob->fg_flag, FNORDAHEAD, relaxed);
+               } else {
+                       os_atomic_or(&fp->fp_glob->fg_flag, FNORDAHEAD, relaxed);
+               }
                 goto out;
  
         case F_NOCACHE:
                 goto out;
  
         case F_NOCACHE:
@@ -1440,11 +1707,11 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         error = EBADF;
                         goto out;
                 }
                         error = EBADF;
                         goto out;
                 }
-               if (uap->arg)
-                       fp->f_fglob->fg_flag |= FNOCACHE;
-               else
-                       fp->f_fglob->fg_flag &= ~FNOCACHE;
-
+               if (uap->arg) {
+                       os_atomic_or(&fp->fp_glob->fg_flag, FNOCACHE, relaxed);
+               } else {
+                       os_atomic_andnot(&fp->fp_glob->fg_flag, FNOCACHE, relaxed);
+               }
                 goto out;
  
         case F_NODIRECT:
                 goto out;
  
         case F_NODIRECT:
@@ -1452,11 +1719,11 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         error = EBADF;
                         goto out;
                 }
                         error = EBADF;
                         goto out;
                 }
-               if (uap->arg)
-                       fp->f_fglob->fg_flag |= FNODIRECT;
-               else
-                       fp->f_fglob->fg_flag &= ~FNODIRECT;
-
+               if (uap->arg) {
+                       os_atomic_or(&fp->fp_glob->fg_flag, FNODIRECT, relaxed);
+               } else {
+                       os_atomic_andnot(&fp->fp_glob->fg_flag, FNODIRECT, relaxed);
+               }
                 goto out;
  
         case F_SINGLE_WRITER:
                 goto out;
  
         case F_SINGLE_WRITER:
@@ -1464,50 +1731,50 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         error = EBADF;
                         goto out;
                 }
                         error = EBADF;
                         goto out;
                 }
-               if (uap->arg)
-                       fp->f_fglob->fg_flag |= FSINGLE_WRITER;
-               else
-                       fp->f_fglob->fg_flag &= ~FSINGLE_WRITER;
-
+               if (uap->arg) {
+                       os_atomic_or(&fp->fp_glob->fg_flag, FSINGLE_WRITER, relaxed);
+               } else {
+                       os_atomic_andnot(&fp->fp_glob->fg_flag, FSINGLE_WRITER, relaxed);
+               }
                 goto out;
  
         case F_GLOBAL_NOCACHE:
                 goto out;
  
         case F_GLOBAL_NOCACHE:
-               if (fp->f_type != DTYPE_VNODE) {
-                       error = EBADF;
+               if (fp->f_type != DTYPE_VNODE) {
+                       error = EBADF;
                         goto out;
                 }
                 vp = (struct vnode *)fp->f_data;
                 proc_fdunlock(p);
  
                         goto out;
                 }
                 vp = (struct vnode *)fp->f_data;
                 proc_fdunlock(p);
  
-               if ( (error = vnode_getwithref(vp)) == 0 ) {
+               if ((error = vnode_getwithref(vp)) == 0) {
+                       *retval = vnode_isnocache(vp);
  
  
-                       *retval = vnode_isnocache(vp);
-
-                       if (uap->arg)
-                               vnode_setnocache(vp);
-                       else
-                               vnode_clearnocache(vp);
+                       if (uap->arg) {
+                               vnode_setnocache(vp);
+                       } else {
+                               vnode_clearnocache(vp);
+                       }
  
                         (void)vnode_put(vp);
                 }
                 goto outdrop;
  
         case F_CHECK_OPENEVT:
  
                         (void)vnode_put(vp);
                 }
                 goto outdrop;
  
         case F_CHECK_OPENEVT:
-               if (fp->f_type != DTYPE_VNODE) {
-                       error = EBADF;
+               if (fp->f_type != DTYPE_VNODE) {
+                       error = EBADF;
                         goto out;
                 }
                 vp = (struct vnode *)fp->f_data;
                 proc_fdunlock(p);
  
                         goto out;
                 }
                 vp = (struct vnode *)fp->f_data;
                 proc_fdunlock(p);
  
-               if ( (error = vnode_getwithref(vp)) == 0 ) {
-
-                       *retval = vnode_is_openevt(vp);
+               if ((error = vnode_getwithref(vp)) == 0) {
+                       *retval = vnode_is_openevt(vp);
  
  
-                       if (uap->arg)
-                               vnode_set_openevt(vp);
-                       else
-                               vnode_clear_openevt(vp);
+                       if (uap->arg) {
+                               vnode_set_openevt(vp);
+                       } else {
+                               vnode_clear_openevt(vp);
+                       }
  
                         (void)vnode_put(vp);
                 }
  
                         (void)vnode_put(vp);
                 }
@@ -1523,31 +1790,36 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 vp = (struct vnode *)fp->f_data;
                 proc_fdunlock(p);
  
                 vp = (struct vnode *)fp->f_data;
                 proc_fdunlock(p);
  
-               if ( (error = copyin(argp, (caddr_t)&ra_struct, sizeof(ra_struct))) )
+               if ((error = copyin(argp, (caddr_t)&ra_struct, sizeof(ra_struct)))) {
                         goto outdrop;
                         goto outdrop;
-               if ( (error = vnode_getwithref(vp)) == 0 ) {
-                       error = VNOP_IOCTL(vp, F_RDADVISE, (caddr_t)&ra_struct, 0, &context);
+               }
+               if (ra_struct.ra_offset < 0 || ra_struct.ra_count < 0) {
+                       error = EINVAL;
+                       goto outdrop;
+               }
+               if ((error = vnode_getwithref(vp)) == 0) {
+                       error = VNOP_IOCTL(vp, F_RDADVISE, (caddr_t)&ra_struct, 0, &context);
  
                         (void)vnode_put(vp);
                 }
                 goto outdrop;
  
                         (void)vnode_put(vp);
                 }
                 goto outdrop;
-               }
+       }
  
  
-        case F_FLUSH_DATA:
+       case F_FLUSH_DATA:
  
  
-                if (fp->f_type != DTYPE_VNODE) {
-                        error = EBADF;
-                        goto out;
-                }
-                vp = (struct vnode *)fp->f_data;
-                proc_fdunlock(p);
+               if (fp->f_type != DTYPE_VNODE) {
+                       error = EBADF;
+                       goto out;
+               }
+               vp = (struct vnode *)fp->f_data;
+               proc_fdunlock(p);
  
  
-                if ( (error = vnode_getwithref(vp)) == 0 ) {
-                        error = VNOP_FSYNC(vp, MNT_NOWAIT, &context);
+               if ((error = vnode_getwithref(vp)) == 0) {
+                       error = VNOP_FSYNC(vp, MNT_NOWAIT, &context);
  
  
-                        (void)vnode_put(vp);
-                }
-                goto outdrop;
+                       (void)vnode_put(vp);
+               }
+               goto outdrop;
  
         case F_LOG2PHYS:
         case F_LOG2PHYS_EXT: {
  
         case F_LOG2PHYS:
         case F_LOG2PHYS_EXT: {
@@ -1560,8 +1832,9 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  
                 if (uap->cmd == F_LOG2PHYS_EXT) {
                         error = copyin(argp, (caddr_t)&l2p_struct, sizeof(l2p_struct));
  
                 if (uap->cmd == F_LOG2PHYS_EXT) {
                         error = copyin(argp, (caddr_t)&l2p_struct, sizeof(l2p_struct));
-                       if (error)
+                       if (error) {
                                 goto out;
                                 goto out;
+                       }
                         file_offset = l2p_struct.l2p_devoffset;
                 } else {
                         file_offset = fp->f_offset;
                         file_offset = l2p_struct.l2p_devoffset;
                 } else {
                         file_offset = fp->f_offset;
@@ -1572,7 +1845,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 }
                 vp = (struct vnode *)fp->f_data;
                 proc_fdunlock(p);
                 }
                 vp = (struct vnode *)fp->f_data;
                 proc_fdunlock(p);
-               if ( (error = vnode_getwithref(vp)) ) {
+               if ((error = vnode_getwithref(vp))) {
                         goto outdrop;
                 }
                 error = VNOP_OFFTOBLK(vp, file_offset, &lbn);
                         goto outdrop;
                 }
                 error = VNOP_OFFTOBLK(vp, file_offset, &lbn);
@@ -1593,7 +1866,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                                 goto outdrop;
                         }
  
                                 goto outdrop;
                         }
  
-                       a_size = MIN((uint64_t)l2p_struct.l2p_contigbytes, SIZE_MAX);
+                       a_size = (size_t)MIN((uint64_t)l2p_struct.l2p_contigbytes, SIZE_MAX);
                 } else {
                         a_size = devBlockSize;
                 }
                 } else {
                         a_size = devBlockSize;
                 }
@@ -1603,11 +1876,11 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 (void)vnode_put(vp);
  
                 if (!error) {
                 (void)vnode_put(vp);
  
                 if (!error) {
-                       l2p_struct.l2p_flags = 0;       /* for now */
+                       l2p_struct.l2p_flags = 0;       /* for now */
                         if (uap->cmd == F_LOG2PHYS_EXT) {
                                 l2p_struct.l2p_contigbytes = run - (file_offset - offset);
                         } else {
                         if (uap->cmd == F_LOG2PHYS_EXT) {
                                 l2p_struct.l2p_contigbytes = run - (file_offset - offset);
                         } else {
-                               l2p_struct.l2p_contigbytes = 0; /* for now */
+                               l2p_struct.l2p_contigbytes = 0; /* for now */
                         }
  
                         /*
                         }
  
                         /*
@@ -1617,16 +1890,16 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         if (bn == -1) {
                                 /* Don't multiply it by the block size */
                                 l2p_struct.l2p_devoffset = bn;
                         if (bn == -1) {
                                 /* Don't multiply it by the block size */
                                 l2p_struct.l2p_devoffset = bn;
-                       }
-                       else {
+                       } else {
                                 l2p_struct.l2p_devoffset = bn * devBlockSize;
                                 l2p_struct.l2p_devoffset += file_offset - offset;
                         }
                         error = copyout((caddr_t)&l2p_struct, argp, sizeof(l2p_struct));
                 }
                 goto outdrop;
                                 l2p_struct.l2p_devoffset = bn * devBlockSize;
                                 l2p_struct.l2p_devoffset += file_offset - offset;
                         }
                         error = copyout((caddr_t)&l2p_struct, argp, sizeof(l2p_struct));
                 }
                 goto outdrop;
-               }
-       case F_GETPATH: {
+       }
+       case F_GETPATH:
+       case F_GETPATH_NOFIRMLINK: {
                 char *pathbufp;
                 int pathlen;
  
                 char *pathbufp;
                 int pathlen;
  
@@ -1638,19 +1911,21 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 proc_fdunlock(p);
  
                 pathlen = MAXPATHLEN;
                 proc_fdunlock(p);
  
                 pathlen = MAXPATHLEN;
-               MALLOC(pathbufp, char *, pathlen, M_TEMP, M_WAITOK);
-               if (pathbufp == NULL) {
-                       error = ENOMEM;
-                       goto outdrop;
-               }
-               if ( (error = vnode_getwithref(vp)) == 0 ) {
-                       error = vn_getpath(vp, pathbufp, &pathlen);
-                       (void)vnode_put(vp);
+               pathbufp = zalloc(ZV_NAMEI);
  
  
-                       if (error == 0)
-                               error = copyout((caddr_t)pathbufp, argp, pathlen);
+               if ((error = vnode_getwithref(vp)) == 0) {
+                       if (uap->cmd == F_GETPATH_NOFIRMLINK) {
+                               error = vn_getpath_ext(vp, NULL, pathbufp, &pathlen, VN_GETPATH_NO_FIRMLINK);
+                       } else {
+                               error = vn_getpath(vp, pathbufp, &pathlen);
+                       }
+                       (void)vnode_put(vp);
+
+                       if (error == 0) {
+                               error = copyout((caddr_t)pathbufp, argp, pathlen);
+                       }
                 }
                 }
-               FREE(pathbufp, M_TEMP);
+               zfree(ZV_NAMEI, pathbufp);
                 goto outdrop;
         }
  
                 goto outdrop;
         }
  
@@ -1659,24 +1934,24 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 size_t pathlen;
  
                 if (fp->f_type != DTYPE_VNODE) {
                 size_t pathlen;
  
                 if (fp->f_type != DTYPE_VNODE) {
-                       error = EBADF;
+                       error = EBADF;
                         goto out;
                 }
                 vp = (struct vnode *)fp->f_data;
                 proc_fdunlock(p);
  
                 pathlen = MAXPATHLEN;
                         goto out;
                 }
                 vp = (struct vnode *)fp->f_data;
                 proc_fdunlock(p);
  
                 pathlen = MAXPATHLEN;
-               pathbufp = kalloc(MAXPATHLEN);
+               pathbufp = zalloc(ZV_NAMEI);
  
  
-               if ( (error = copyinstr(argp, pathbufp, MAXPATHLEN, &pathlen)) == 0 ) {
-                       if ( (error = vnode_getwithref(vp)) == 0 ) {
+               if ((error = copyinstr(argp, pathbufp, MAXPATHLEN, &pathlen)) == 0) {
+                       if ((error = vnode_getwithref(vp)) == 0) {
                                 AUDIT_ARG(text, pathbufp);
                                 AUDIT_ARG(text, pathbufp);
-                               error = vn_path_package_check(vp, pathbufp, pathlen, retval);
+                               error = vn_path_package_check(vp, pathbufp, (int)pathlen, retval);
  
                                 (void)vnode_put(vp);
                         }
                 }
  
                                 (void)vnode_put(vp);
                         }
                 }
-               kfree(pathbufp, MAXPATHLEN);
+               zfree(ZV_NAMEI, pathbufp);
                 goto outdrop;
         }
  
                 goto outdrop;
         }
  
@@ -1692,8 +1967,8 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 vp = (struct vnode *)fp->f_data;
                 proc_fdunlock(p);
  
                 vp = (struct vnode *)fp->f_data;
                 proc_fdunlock(p);
  
-               if ( (error = vnode_getwithref(vp)) == 0 ) {
-                       error = VNOP_IOCTL(vp, uap->cmd, (caddr_t)NULL, 0, &context);
+               if ((error = vnode_getwithref(vp)) == 0) {
+                       error = VNOP_IOCTL(vp, uap->cmd, (caddr_t)NULL, 0, &context);
  
                         (void)vnode_put(vp);
                 }
  
                         (void)vnode_put(vp);
                 }
@@ -1730,6 +2005,15 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         goto outdrop;
                 }
  
                         goto outdrop;
                 }
  
+               /*
+                * Only entitled apps may use the credentials of the thread
+                * that opened the file descriptor.
+                * Non-entitled threads will use their own context.
+                */
+               if (IOTaskHasEntitlement(current_task(), ACCOUNT_OPENFROM_ENTITLEMENT)) {
+                       has_entitlement = 1;
+               }
+
                 /* Get flags, mode and pathname arguments. */
                 if (IS_64BIT_PROCESS(p)) {
                         error = copyin(argp, &fopen, sizeof(fopen));
                 /* Get flags, mode and pathname arguments. */
                 if (IS_64BIT_PROCESS(p)) {
                         error = copyin(argp, &fopen, sizeof(fopen));
@@ -1749,16 +2033,16 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 AUDIT_ARG(mode, fopen.o_mode);
                 VATTR_INIT(&va);
                 /* Mask off all but regular access permissions */
                 AUDIT_ARG(mode, fopen.o_mode);
                 VATTR_INIT(&va);
                 /* Mask off all but regular access permissions */
-               cmode = ((fopen.o_mode &~ fdp->fd_cmask) & ALLPERMS) & ~S_ISTXT;
+               cmode = ((fopen.o_mode & ~fdp->fd_cmask) & ALLPERMS) & ~S_ISTXT;
                 VATTR_SET(&va, va_mode, cmode & ACCESSPERMS);
  
                 /* Start the lookup relative to the file descriptor's vnode. */
                 NDINIT(&nd, LOOKUP, OP_OPEN, USEDVP | FOLLOW | AUDITVNPATH1, UIO_USERSPACE,
                 VATTR_SET(&va, va_mode, cmode & ACCESSPERMS);
  
                 /* Start the lookup relative to the file descriptor's vnode. */
                 NDINIT(&nd, LOOKUP, OP_OPEN, USEDVP | FOLLOW | AUDITVNPATH1, UIO_USERSPACE,
-                      fopen.o_pathname, &context);
+                   fopen.o_pathname, has_entitlement ? &context : vfs_context_current());
                 nd.ni_dvp = vp;
  
                 nd.ni_dvp = vp;
  
-               error = open1(&context, &nd, fopen.o_flags, &va,
-                             fileproc_alloc_init, NULL, retval);
+               error = open1(has_entitlement ? &context : vfs_context_current(),
+                   &nd, fopen.o_flags, &va, fileproc_alloc_init, NULL, retval);
  
                 vnode_put(vp);
                 break;
  
                 vnode_put(vp);
                 break;
@@ -1790,6 +2074,15 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         goto outdrop;
                 }
  
                         goto outdrop;
                 }
  
+               /*
+                * Only entitled apps may use the credentials of the thread
+                * that opened the file descriptor.
+                * Non-entitled threads will use their own context.
+                */
+               if (IOTaskHasEntitlement(current_task(), ACCOUNT_OPENFROM_ENTITLEMENT)) {
+                       has_entitlement = 1;
+               }
+
                 /* Get flags, mode and pathname arguments. */
                 if (IS_64BIT_PROCESS(p)) {
                         pathname = (user_addr_t)argp;
                 /* Get flags, mode and pathname arguments. */
                 if (IS_64BIT_PROCESS(p)) {
                         pathname = (user_addr_t)argp;
@@ -1798,17 +2091,18 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 }
  
                 /* Start the lookup relative to the file descriptor's vnode. */
                 }
  
                 /* Start the lookup relative to the file descriptor's vnode. */
-               error = unlink1(&context, vp, pathname, UIO_USERSPACE, 0);
+               error = unlink1(has_entitlement ? &context : vfs_context_current(),
+                   vp, pathname, UIO_USERSPACE, 0);
  
                 vnode_put(vp);
                 break;
  
                 vnode_put(vp);
                 break;
-
         }
  
         case F_ADDSIGS:
         case F_ADDFILESIGS:
         case F_ADDFILESIGS_FOR_DYLD_SIM:
         case F_ADDFILESIGS_RETURN:
         }
  
         case F_ADDSIGS:
         case F_ADDFILESIGS:
         case F_ADDFILESIGS_FOR_DYLD_SIM:
         case F_ADDFILESIGS_RETURN:
+       case F_ADDFILESIGS_INFO:
         {
                 struct cs_blob *blob = NULL;
                 struct user_fsignatures fs;
         {
                 struct cs_blob *blob = NULL;
                 struct user_fsignatures fs;
@@ -1816,6 +2110,9 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 vm_offset_t kernel_blob_addr;
                 vm_size_t kernel_blob_size;
                 int blob_add_flags = 0;
                 vm_offset_t kernel_blob_addr;
                 vm_size_t kernel_blob_size;
                 int blob_add_flags = 0;
+               const size_t sizeof_fs = (uap->cmd == F_ADDFILESIGS_INFO ?
+                   offsetof(struct user_fsignatures, fs_cdhash /* first output element */) :
+                   offsetof(struct user_fsignatures, fs_fsignatures_size /* compat */));
  
                 if (fp->f_type != DTYPE_VNODE) {
                         error = EBADF;
  
                 if (fp->f_type != DTYPE_VNODE) {
                         error = EBADF;
@@ -1834,15 +2131,22 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 }
  
                 error = vnode_getwithref(vp);
                 }
  
                 error = vnode_getwithref(vp);
-               if (error)
+               if (error) {
                         goto outdrop;
                         goto outdrop;
+               }
  
                 if (IS_64BIT_PROCESS(p)) {
  
                 if (IS_64BIT_PROCESS(p)) {
-                       error = copyin(argp, &fs, sizeof (fs));
+                       error = copyin(argp, &fs, sizeof_fs);
                 } else {
                 } else {
+                       if (uap->cmd == F_ADDFILESIGS_INFO) {
+                               error = EINVAL;
+                               vnode_put(vp);
+                               goto outdrop;
+                       }
+
                         struct user32_fsignatures fs32;
  
                         struct user32_fsignatures fs32;
  
-                       error = copyin(argp, &fs32, sizeof (fs32));
+                       error = copyin(argp, &fs32, sizeof(fs32));
                         fs.fs_file_start = fs32.fs_file_start;
                         fs.fs_blob_start = CAST_USER_ADDR_T(fs32.fs_blob_start);
                         fs.fs_blob_size = fs32.fs_blob_size;
                         fs.fs_file_start = fs32.fs_file_start;
                         fs.fs_blob_start = CAST_USER_ADDR_T(fs32.fs_blob_start);
                         fs.fs_blob_size = fs32.fs_blob_size;
@@ -1856,12 +2160,11 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 /*
                  * First check if we have something loaded a this offset
                  */
                 /*
                  * First check if we have something loaded a this offset
                  */
-               blob = ubc_cs_blob_get(vp, CPU_TYPE_ANY, fs.fs_file_start);
-               if (blob != NULL)
-               {
+               blob = ubc_cs_blob_get(vp, CPU_TYPE_ANY, CPU_SUBTYPE_ANY, fs.fs_file_start);
+               if (blob != NULL) {
                         /* If this is for dyld_sim revalidate the blob */
                         if (uap->cmd == F_ADDFILESIGS_FOR_DYLD_SIM) {
                         /* If this is for dyld_sim revalidate the blob */
                         if (uap->cmd == F_ADDFILESIGS_FOR_DYLD_SIM) {
-                               error = ubc_cs_blob_revalidate(vp, blob, NULL, blob_add_flags);
+                               error = ubc_cs_blob_revalidate(vp, blob, NULL, blob_add_flags, proc_platform(p));
                                 if (error) {
                                         blob = NULL;
                                         if (error != EAGAIN) {
                                 if (error) {
                                         blob = NULL;
                                         if (error != EAGAIN) {
@@ -1891,29 +2194,29 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  
                         kernel_blob_size = CAST_DOWN(vm_size_t, fs.fs_blob_size);
                         kr = ubc_cs_blob_allocate(&kernel_blob_addr, &kernel_blob_size);
  
                         kernel_blob_size = CAST_DOWN(vm_size_t, fs.fs_blob_size);
                         kr = ubc_cs_blob_allocate(&kernel_blob_addr, &kernel_blob_size);
-                       if (kr != KERN_SUCCESS) {
+                       if (kr != KERN_SUCCESS || kernel_blob_size < fs.fs_blob_size) {
                                 error = ENOMEM;
                                 vnode_put(vp);
                                 goto outdrop;
                         }
  
                                 error = ENOMEM;
                                 vnode_put(vp);
                                 goto outdrop;
                         }
  
-                       if(uap->cmd == F_ADDSIGS) {
+                       if (uap->cmd == F_ADDSIGS) {
                                 error = copyin(fs.fs_blob_start,
                                 error = copyin(fs.fs_blob_start,
-                                              (void *) kernel_blob_addr,
-                                              kernel_blob_size);
-                       } else /* F_ADDFILESIGS || F_ADDFILESIGS_RETURN || F_ADDFILESIGS_FOR_DYLD_SIM */ {
+                                   (void *) kernel_blob_addr,
+                                   fs.fs_blob_size);
+                       } else { /* F_ADDFILESIGS || F_ADDFILESIGS_RETURN || F_ADDFILESIGS_FOR_DYLD_SIM || F_ADDFILESIGS_INFO */
                                 int resid;
  
                                 error = vn_rdwr(UIO_READ,
                                 int resid;
  
                                 error = vn_rdwr(UIO_READ,
-                                               vp,
-                                               (caddr_t) kernel_blob_addr,
-                                               kernel_blob_size,
-                                               fs.fs_file_start + fs.fs_blob_start,
-                                               UIO_SYSSPACE,
-                                               0,
-                                               kauth_cred_get(),
-                                               &resid,
-                                               p);
+                                   vp,
+                                   (caddr_t) kernel_blob_addr,
+                                   (int)kernel_blob_size,
+                                   fs.fs_file_start + fs.fs_blob_start,
+                                   UIO_SYSSPACE,
+                                   0,
+                                   kauth_cred_get(),
+                                   &resid,
+                                   p);
                                 if ((error == 0) && resid) {
                                         /* kernel_blob_size rounded to a page size, but signature may be at end of file */
                                         memset((void *)(kernel_blob_addr + (kernel_blob_size - resid)), 0x0, resid);
                                 if ((error == 0) && resid) {
                                         /* kernel_blob_size rounded to a page size, but signature may be at end of file */
                                         memset((void *)(kernel_blob_addr + (kernel_blob_size - resid)), 0x0, resid);
@@ -1922,26 +2225,28 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  
                         if (error) {
                                 ubc_cs_blob_deallocate(kernel_blob_addr,
  
                         if (error) {
                                 ubc_cs_blob_deallocate(kernel_blob_addr,
-                                                      kernel_blob_size);
+                                   kernel_blob_size);
                                 vnode_put(vp);
                                 goto outdrop;
                         }
  
                         blob = NULL;
                         error = ubc_cs_blob_add(vp,
                                 vnode_put(vp);
                                 goto outdrop;
                         }
  
                         blob = NULL;
                         error = ubc_cs_blob_add(vp,
-                                               CPU_TYPE_ANY,   /* not for a specific architecture */
-                                               fs.fs_file_start,
-                                               &kernel_blob_addr,
-                                               kernel_blob_size,
-                                               NULL,
-                                               blob_add_flags,
-                                               &blob);
+                           proc_platform(p),
+                           CPU_TYPE_ANY,                       /* not for a specific architecture */
+                           CPU_SUBTYPE_ANY,
+                           fs.fs_file_start,
+                           &kernel_blob_addr,
+                           kernel_blob_size,
+                           NULL,
+                           blob_add_flags,
+                           &blob);
  
                         /* ubc_blob_add() has consumed "kernel_blob_addr" if it is zeroed */
                         if (error) {
                                 if (kernel_blob_addr) {
                                         ubc_cs_blob_deallocate(kernel_blob_addr,
  
                         /* ubc_blob_add() has consumed "kernel_blob_addr" if it is zeroed */
                         if (error) {
                                 if (kernel_blob_addr) {
                                         ubc_cs_blob_deallocate(kernel_blob_addr,
-                                                              kernel_blob_size);
+                                           kernel_blob_size);
                                 }
                                 vnode_put(vp);
                                 goto outdrop;
                                 }
                                 vnode_put(vp);
                                 goto outdrop;
@@ -1952,21 +2257,165 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         }
                 }
  
                         }
                 }
  
-               if (uap->cmd == F_ADDFILESIGS_RETURN || uap->cmd == F_ADDFILESIGS_FOR_DYLD_SIM) {
+               if (uap->cmd == F_ADDFILESIGS_RETURN || uap->cmd == F_ADDFILESIGS_FOR_DYLD_SIM ||
+                   uap->cmd == F_ADDFILESIGS_INFO) {
                         /*
                          * The first element of the structure is a
                          * off_t that happen to have the same size for
                          * all archs. Lets overwrite that.
                          */
                         off_t end_offset = 0;
                         /*
                          * The first element of the structure is a
                          * off_t that happen to have the same size for
                          * all archs. Lets overwrite that.
                          */
                         off_t end_offset = 0;
-                       if (blob)
+                       if (blob) {
                                 end_offset = blob->csb_end_offset;
                                 end_offset = blob->csb_end_offset;
-                       error = copyout(&end_offset, argp, sizeof (end_offset));
+                       }
+                       error = copyout(&end_offset, argp, sizeof(end_offset));
+
+                       if (error) {
+                               vnode_put(vp);
+                               goto outdrop;
+                       }
+               }
+
+               if (uap->cmd == F_ADDFILESIGS_INFO) {
+                       /* Return information. What we copy out depends on the size of the
+                        * passed in structure, to keep binary compatibility. */
+
+                       if (fs.fs_fsignatures_size >= sizeof(struct user_fsignatures)) {
+                               // enough room for fs_cdhash[20]+fs_hash_type
+
+                               if (blob != NULL) {
+                                       error = copyout(blob->csb_cdhash,
+                                           (vm_address_t)argp + offsetof(struct user_fsignatures, fs_cdhash),
+                                           USER_FSIGNATURES_CDHASH_LEN);
+                                       if (error) {
+                                               vnode_put(vp);
+                                               goto outdrop;
+                                       }
+                                       int hashtype = cs_hash_type(blob->csb_hashtype);
+                                       error = copyout(&hashtype,
+                                           (vm_address_t)argp + offsetof(struct user_fsignatures, fs_hash_type),
+                                           sizeof(int));
+                                       if (error) {
+                                               vnode_put(vp);
+                                               goto outdrop;
+                                       }
+                               }
+                       }
                 }
  
                 (void) vnode_put(vp);
                 break;
         }
                 }
  
                 (void) vnode_put(vp);
                 break;
         }
+#if CONFIG_SUPPLEMENTAL_SIGNATURES
+       case F_ADDFILESUPPL:
+       {
+               struct vnode *ivp;
+               struct cs_blob *blob = NULL;
+               struct user_fsupplement fs;
+               int orig_fd;
+               struct fileproc* orig_fp = NULL;
+               kern_return_t kr;
+               vm_offset_t kernel_blob_addr;
+               vm_size_t kernel_blob_size;
+
+               if (!IS_64BIT_PROCESS(p)) {
+                       error = EINVAL;
+                       goto out; // drop fp and unlock fds
+               }
+
+               if (fp->f_type != DTYPE_VNODE) {
+                       error = EBADF;
+                       goto out;
+               }
+
+               error = copyin(argp, &fs, sizeof(fs));
+               if (error) {
+                       goto out;
+               }
+
+               orig_fd = fs.fs_orig_fd;
+               if ((error = fp_lookup(p, orig_fd, &orig_fp, 1))) {
+                       printf("CODE SIGNING: Failed to find original file for supplemental signature attachment\n");
+                       goto out;
+               }
+
+               if (orig_fp->f_type != DTYPE_VNODE) {
+                       error = EBADF;
+                       fp_drop(p, orig_fd, orig_fp, 1);
+                       goto out;
+               }
+
+               ivp = (struct vnode *)orig_fp->f_data;
+
+               vp = (struct vnode *)fp->f_data;
+
+               proc_fdunlock(p);
+
+               error = vnode_getwithref(ivp);
+               if (error) {
+                       fp_drop(p, orig_fd, orig_fp, 0);
+                       goto outdrop; //drop fp
+               }
+
+               error = vnode_getwithref(vp);
+               if (error) {
+                       vnode_put(ivp);
+                       fp_drop(p, orig_fd, orig_fp, 0);
+                       goto outdrop;
+               }
+
+               if (fs.fs_blob_size > CS_MAX_BLOB_SIZE) {
+                       error = E2BIG;
+                       goto dropboth; // drop iocounts on vp and ivp, drop orig_fp then drop fp via outdrop
+               }
+
+               kernel_blob_size = CAST_DOWN(vm_size_t, fs.fs_blob_size);
+               kr = ubc_cs_blob_allocate(&kernel_blob_addr, &kernel_blob_size);
+               if (kr != KERN_SUCCESS) {
+                       error = ENOMEM;
+                       goto dropboth;
+               }
+
+               int resid;
+               error = vn_rdwr(UIO_READ, vp,
+                   (caddr_t)kernel_blob_addr, (int)kernel_blob_size,
+                   fs.fs_file_start + fs.fs_blob_start,
+                   UIO_SYSSPACE, 0,
+                   kauth_cred_get(), &resid, p);
+               if ((error == 0) && resid) {
+                       /* kernel_blob_size rounded to a page size, but signature may be at end of file */
+                       memset((void *)(kernel_blob_addr + (kernel_blob_size - resid)), 0x0, resid);
+               }
+
+               if (error) {
+                       ubc_cs_blob_deallocate(kernel_blob_addr,
+                           kernel_blob_size);
+                       goto dropboth;
+               }
+
+               error = ubc_cs_blob_add_supplement(vp, ivp, fs.fs_file_start,
+                   &kernel_blob_addr, kernel_blob_size, &blob);
+
+               /* ubc_blob_add_supplement() has consumed kernel_blob_addr if it is zeroed */
+               if (error) {
+                       if (kernel_blob_addr) {
+                               ubc_cs_blob_deallocate(kernel_blob_addr,
+                                   kernel_blob_size);
+                       }
+                       goto dropboth;
+               }
+               vnode_put(ivp);
+               vnode_put(vp);
+               fp_drop(p, orig_fd, orig_fp, 0);
+               break;
+
+dropboth:
+               vnode_put(ivp);
+               vnode_put(vp);
+               fp_drop(p, orig_fd, orig_fp, 0);
+               goto outdrop;
+       }
+#endif
         case F_GETCODEDIR:
         case F_FINDSIGS: {
                 error = ENOTSUP;
         case F_GETCODEDIR:
         case F_FINDSIGS: {
                 error = ENOTSUP;
@@ -1974,27 +2423,28 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
         }
         case F_CHECK_LV: {
                 struct fileglob *fg;
         }
         case F_CHECK_LV: {
                 struct fileglob *fg;
-               fchecklv_t lv;
+               fchecklv_t lv = {};
  
                 if (fp->f_type != DTYPE_VNODE) {
                         error = EBADF;
                         goto out;
                 }
  
                 if (fp->f_type != DTYPE_VNODE) {
                         error = EBADF;
                         goto out;
                 }
-               fg = fp->f_fglob;
+               fg = fp->fp_glob;
                 proc_fdunlock(p);
  
                 if (IS_64BIT_PROCESS(p)) {
                 proc_fdunlock(p);
  
                 if (IS_64BIT_PROCESS(p)) {
-                       error = copyin(argp, &lv, sizeof (lv));
+                       error = copyin(argp, &lv, sizeof(lv));
                 } else {
                 } else {
-                       struct user32_fchecklv lv32;
+                       struct user32_fchecklv lv32 = {};
  
  
-                       error = copyin(argp, &lv32, sizeof (lv32));
+                       error = copyin(argp, &lv32, sizeof(lv32));
                         lv.lv_file_start = lv32.lv_file_start;
                         lv.lv_error_message = (void *)(uintptr_t)lv32.lv_error_message;
                         lv.lv_file_start = lv32.lv_file_start;
                         lv.lv_error_message = (void *)(uintptr_t)lv32.lv_error_message;
-                       lv.lv_error_message_size = lv32.lv_error_message;
+                       lv.lv_error_message_size = lv32.lv_error_message_size;
                 }
                 }
-               if (error)
+               if (error) {
                         goto outdrop;
                         goto outdrop;
+               }
  
  #if CONFIG_MACF
                 error = mac_file_check_library_validation(p, fg, lv.lv_file_start,
  
  #if CONFIG_MACF
                 error = mac_file_check_library_validation(p, fg, lv.lv_file_start,
@@ -2003,6 +2453,53 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  
                 break;
         }
  
                 break;
         }
+       case F_GETSIGSINFO: {
+               struct cs_blob *blob = NULL;
+               fgetsigsinfo_t sigsinfo = {};
+
+               if (fp->f_type != DTYPE_VNODE) {
+                       error = EBADF;
+                       goto out;
+               }
+               vp = (struct vnode *)fp->f_data;
+               proc_fdunlock(p);
+
+               error = vnode_getwithref(vp);
+               if (error) {
+                       goto outdrop;
+               }
+
+               error = copyin(argp, &sigsinfo, sizeof(sigsinfo));
+               if (error) {
+                       vnode_put(vp);
+                       goto outdrop;
+               }
+
+               blob = ubc_cs_blob_get(vp, CPU_TYPE_ANY, CPU_SUBTYPE_ANY, sigsinfo.fg_file_start);
+               if (blob == NULL) {
+                       error = ENOENT;
+                       vnode_put(vp);
+                       goto outdrop;
+               }
+               switch (sigsinfo.fg_info_request) {
+               case GETSIGSINFO_PLATFORM_BINARY:
+                       sigsinfo.fg_sig_is_platform = blob->csb_platform_binary;
+                       error = copyout(&sigsinfo.fg_sig_is_platform,
+                           (vm_address_t)argp + offsetof(struct fgetsigsinfo, fg_sig_is_platform),
+                           sizeof(sigsinfo.fg_sig_is_platform));
+                       if (error) {
+                               vnode_put(vp);
+                               goto outdrop;
+                       }
+                       break;
+               default:
+                       error = EINVAL;
+                       vnode_put(vp);
+                       goto outdrop;
+               }
+               vnode_put(vp);
+               break;
+       }
  #if CONFIG_PROTECT
         case F_GETPROTECTIONCLASS: {
                 if (fp->f_type != DTYPE_VNODE) {
  #if CONFIG_PROTECT
         case F_GETPROTECTIONCLASS: {
                 if (fp->f_type != DTYPE_VNODE) {
@@ -2024,10 +2521,11 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 VATTR_WANTED(&va, va_dataprotect_class);
                 error = VNOP_GETATTR(vp, &va, &context);
                 if (!error) {
                 VATTR_WANTED(&va, va_dataprotect_class);
                 error = VNOP_GETATTR(vp, &va, &context);
                 if (!error) {
-                       if (VATTR_IS_SUPPORTED(&va, va_dataprotect_class))
+                       if (VATTR_IS_SUPPORTED(&va, va_dataprotect_class)) {
                                 *retval = va.va_dataprotect_class;
                                 *retval = va.va_dataprotect_class;
-                       else
+                       } else {
                                 error = ENOTSUP;
                                 error = ENOTSUP;
+                       }
                 }
  
                 vnode_put(vp);
                 }
  
                 vnode_put(vp);
@@ -2053,7 +2551,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  
                 /* Only go forward if you have write access */
                 vfs_context_t ctx = vfs_context_current();
  
                 /* Only go forward if you have write access */
                 vfs_context_t ctx = vfs_context_current();
-               if(vnode_authorize(vp, NULLVP, (KAUTH_VNODE_ACCESS | KAUTH_VNODE_WRITE_DATA), ctx) != 0) {
+               if (vnode_authorize(vp, NULLVP, (KAUTH_VNODE_ACCESS | KAUTH_VNODE_WRITE_DATA), ctx) != 0) {
                         vnode_put(vp);
                         error = EBADF;
                         goto outdrop;
                         vnode_put(vp);
                         error = EBADF;
                         goto outdrop;
@@ -2088,9 +2586,12 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         .len = CP_MAX_WRAPPEDKEYSIZE,
                 };
  
                         .len = CP_MAX_WRAPPEDKEYSIZE,
                 };
  
-               MALLOC(k.key, char *, k.len, M_TEMP, M_WAITOK | M_ZERO);
-
-               error = VNOP_IOCTL(vp, F_TRANSCODEKEY, (caddr_t)&k, 1, &context);
+               k.key = kheap_alloc(KHEAP_TEMP, CP_MAX_WRAPPEDKEYSIZE, Z_WAITOK | Z_ZERO);
+               if (k.key == NULL) {
+                       error = ENOMEM;
+               } else {
+                       error = VNOP_IOCTL(vp, F_TRANSCODEKEY, (caddr_t)&k, 1, &context);
+               }
  
                 vnode_put(vp);
  
  
                 vnode_put(vp);
  
@@ -2099,7 +2600,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         *retval = k.len;
                 }
  
                         *retval = k.len;
                 }
  
-               FREE(k.key, M_TEMP);
+               kheap_free(KHEAP_TEMP, k.key, CP_MAX_WRAPPEDKEYSIZE);
  
                 break;
         }
  
                 break;
         }
@@ -2111,7 +2612,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 }
  
                 vp = (struct vnode*) fp->f_data;
                 }
  
                 vp = (struct vnode*) fp->f_data;
-               proc_fdunlock (p);
+               proc_fdunlock(p);
  
                 if (vnode_getwithref(vp)) {
                         error = ENOENT;
  
                 if (vnode_getwithref(vp)) {
                         error = ENOENT;
@@ -2120,7 +2621,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  
                 error = VNOP_IOCTL(vp, F_GETPROTECTIONLEVEL, (caddr_t)retval, 0, &context);
  
  
                 error = VNOP_IOCTL(vp, F_GETPROTECTIONLEVEL, (caddr_t)retval, 0, &context);
  
-               vnode_put (vp);
+               vnode_put(vp);
                 break;
         }
  
                 break;
         }
  
@@ -2131,7 +2632,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 }
  
                 vp = (struct vnode*) fp->f_data;
                 }
  
                 vp = (struct vnode*) fp->f_data;
-               proc_fdunlock (p);
+               proc_fdunlock(p);
  
                 if (vnode_getwithref(vp)) {
                         error = ENOENT;
  
                 if (vnode_getwithref(vp)) {
                         error = ENOENT;
@@ -2145,7 +2646,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  
                 error = VNOP_IOCTL(vp, F_GETDEFAULTPROTLEVEL, (caddr_t)retval, 0, &context);
  
  
                 error = VNOP_IOCTL(vp, F_GETDEFAULTPROTLEVEL, (caddr_t)retval, 0, &context);
  
-               vnode_put (vp);
+               vnode_put(vp);
                 break;
         }
  
                 break;
         }
  
@@ -2159,17 +2660,21 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 int fd2 = CAST_DOWN_EXPLICIT(int32_t, uap->arg);
  
                 error = priv_check_cred(kauth_cred_get(), PRIV_VFS_MOVE_DATA_EXTENTS, 0);
                 int fd2 = CAST_DOWN_EXPLICIT(int32_t, uap->arg);
  
                 error = priv_check_cred(kauth_cred_get(), PRIV_VFS_MOVE_DATA_EXTENTS, 0);
-               if (error)
+               if (error) {
                         goto out;
                         goto out;
+               }
  
                 if (fp->f_type != DTYPE_VNODE) {
                         error = EBADF;
                         goto out;
                 }
  
  
                 if (fp->f_type != DTYPE_VNODE) {
                         error = EBADF;
                         goto out;
                 }
  
-               /* For now, special case HFS+ only, since this is SPI. */
+               /*
+                * For now, special case HFS+ and APFS only, since this
+                * is SPI.
+                */
                 src_vp = (struct vnode *)fp->f_data;
                 src_vp = (struct vnode *)fp->f_data;
-               if (src_vp->v_tag != VT_HFS) {
+               if (src_vp->v_tag != VT_HFS && src_vp->v_tag != VT_APFS) {
                         error = ENOTSUP;
                         goto out;
                 }
                         error = ENOTSUP;
                         goto out;
                 }
@@ -2178,7 +2683,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                  * Get the references before we start acquiring iocounts on the vnodes,
                  * while we still hold the proc fd lock
                  */
                  * Get the references before we start acquiring iocounts on the vnodes,
                  * while we still hold the proc fd lock
                  */
-               if ( (error = fp_lookup(p, fd2, &fp2, 1)) ) {
+               if ((error = fp_lookup(p, fd2, &fp2, 1))) {
                         error = EBADF;
                         goto out;
                 }
                         error = EBADF;
                         goto out;
                 }
@@ -2188,7 +2693,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         goto out;
                 }
                 dst_vp = (struct vnode *)fp2->f_data;
                         goto out;
                 }
                 dst_vp = (struct vnode *)fp2->f_data;
-               if (dst_vp->v_tag != VT_HFS) {
+               if (dst_vp->v_tag != VT_HFS && dst_vp->v_tag != VT_APFS) {
                         fp_drop(p, fd2, fp2, 1);
                         error = ENOTSUP;
                         goto out;
                         fp_drop(p, fd2, fp2, 1);
                         error = ENOTSUP;
                         goto out;
@@ -2196,7 +2701,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  
  #if CONFIG_MACF
                 /* Re-do MAC checks against the new FD, pass in a fake argument */
  
  #if CONFIG_MACF
                 /* Re-do MAC checks against the new FD, pass in a fake argument */
-               error = mac_file_check_fcntl(proc_ucred(p), fp2->f_fglob, uap->cmd, 0);
+               error = mac_file_check_fcntl(proc_ucred(p), fp2->fp_glob, uap->cmd, 0);
                 if (error) {
                         fp_drop(p, fd2, fp2, 1);
                         goto out;
                 if (error) {
                         fp_drop(p, fd2, fp2, 1);
                         goto out;
@@ -2213,7 +2718,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         goto outdrop;
                 }
                 if (vnode_getwithref(dst_vp)) {
                         goto outdrop;
                 }
                 if (vnode_getwithref(dst_vp)) {
-                       vnode_put (src_vp);
+                       vnode_put(src_vp);
                         fp_drop(p, fd2, fp2, 0);
                         error = ENOENT;
                         goto outdrop;
                         fp_drop(p, fd2, fp2, 0);
                         error = ENOENT;
                         goto outdrop;
@@ -2224,17 +2729,17 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                  * both live on the same filesystem.
                  */
                 if (dst_vp == src_vp) {
                  * both live on the same filesystem.
                  */
                 if (dst_vp == src_vp) {
-                       vnode_put (src_vp);
-                       vnode_put (dst_vp);
-                       fp_drop (p, fd2, fp2, 0);
-                       error = EINVAL;
-                       goto outdrop;
+                       vnode_put(src_vp);
+                       vnode_put(dst_vp);
+                       fp_drop(p, fd2, fp2, 0);
+                       error = EINVAL;
+                       goto outdrop;
                 }
  
                 if (dst_vp->v_mount != src_vp->v_mount) {
                 }
  
                 if (dst_vp->v_mount != src_vp->v_mount) {
-                       vnode_put (src_vp);
-                       vnode_put (dst_vp);
-                       fp_drop (p, fd2, fp2, 0);
+                       vnode_put(src_vp);
+                       vnode_put(dst_vp);
+                       fp_drop(p, fd2, fp2, 0);
                         error = EXDEV;
                         goto outdrop;
                 }
                         error = EXDEV;
                         goto outdrop;
                 }
@@ -2242,8 +2747,8 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 /* Now we have a legit pair of FDs.  Go to work */
  
                 /* Now check for write access to the target files */
                 /* Now we have a legit pair of FDs.  Go to work */
  
                 /* Now check for write access to the target files */
-               if(vnode_authorize(src_vp, NULLVP,
-                                                  (KAUTH_VNODE_ACCESS | KAUTH_VNODE_WRITE_DATA), &context) != 0) {
+               if (vnode_authorize(src_vp, NULLVP,
+                   (KAUTH_VNODE_ACCESS | KAUTH_VNODE_WRITE_DATA), &context) != 0) {
                         vnode_put(src_vp);
                         vnode_put(dst_vp);
                         fp_drop(p, fd2, fp2, 0);
                         vnode_put(src_vp);
                         vnode_put(dst_vp);
                         fp_drop(p, fd2, fp2, 0);
@@ -2251,8 +2756,8 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         goto outdrop;
                 }
  
                         goto outdrop;
                 }
  
-               if(vnode_authorize(dst_vp, NULLVP,
-                                                  (KAUTH_VNODE_ACCESS | KAUTH_VNODE_WRITE_DATA), &context) != 0) {
+               if (vnode_authorize(dst_vp, NULLVP,
+                   (KAUTH_VNODE_ACCESS | KAUTH_VNODE_WRITE_DATA), &context) != 0) {
                         vnode_put(src_vp);
                         vnode_put(dst_vp);
                         fp_drop(p, fd2, fp2, 0);
                         vnode_put(src_vp);
                         vnode_put(dst_vp);
                         fp_drop(p, fd2, fp2, 0);
@@ -2261,11 +2766,11 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 }
  
                 /* Verify that both vps point to files and not directories */
                 }
  
                 /* Verify that both vps point to files and not directories */
-               if ( !vnode_isreg(src_vp) || !vnode_isreg(dst_vp)) {
+               if (!vnode_isreg(src_vp) || !vnode_isreg(dst_vp)) {
                         error = EINVAL;
                         error = EINVAL;
-                       vnode_put (src_vp);
-                       vnode_put (dst_vp);
-                       fp_drop (p, fd2, fp2, 0);
+                       vnode_put(src_vp);
+                       vnode_put(dst_vp);
+                       fp_drop(p, fd2, fp2, 0);
                         goto outdrop;
                 }
  
                         goto outdrop;
                 }
  
@@ -2276,8 +2781,8 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  
                 error = VNOP_EXCHANGE(src_vp, dst_vp, FSOPT_EXCHANGE_DATA_ONLY, &context);
  
  
                 error = VNOP_EXCHANGE(src_vp, dst_vp, FSOPT_EXCHANGE_DATA_ONLY, &context);
  
-               vnode_put (src_vp);
-               vnode_put (dst_vp);
+               vnode_put(src_vp);
+               vnode_put(dst_vp);
                 fp_drop(p, fd2, fp2, 0);
                 break;
         }
                 fp_drop(p, fd2, fp2, 0);
                 break;
         }
@@ -2294,7 +2799,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 }
  
                 vp = (struct vnode*) fp->f_data;
                 }
  
                 vp = (struct vnode*) fp->f_data;
-               proc_fdunlock (p);
+               proc_fdunlock(p);
  
                 /* get the vnode */
                 if (vnode_getwithref(vp)) {
  
                 /* get the vnode */
                 if (vnode_getwithref(vp)) {
@@ -2312,7 +2817,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 /* invoke ioctl to pass off to FS */
                 /* Only go forward if you have write access */
                 vfs_context_t ctx = vfs_context_current();
                 /* invoke ioctl to pass off to FS */
                 /* Only go forward if you have write access */
                 vfs_context_t ctx = vfs_context_current();
-               if(vnode_authorize(vp, NULLVP, (KAUTH_VNODE_ACCESS | KAUTH_VNODE_WRITE_DATA), ctx) != 0) {
+               if (vnode_authorize(vp, NULLVP, (KAUTH_VNODE_ACCESS | KAUTH_VNODE_WRITE_DATA), ctx) != 0) {
                         vnode_put(vp);
                         error = EBADF;
                         goto outdrop;
                         vnode_put(vp);
                         error = EBADF;
                         goto outdrop;
@@ -2320,7 +2825,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  
                 error = VNOP_IOCTL(vp, uap->cmd, (caddr_t)&gcounter, 0, &context);
  
  
                 error = VNOP_IOCTL(vp, uap->cmd, (caddr_t)&gcounter, 0, &context);
  
-               vnode_put (vp);
+               vnode_put(vp);
                 break;
         }
  
                 break;
         }
  
@@ -2329,9 +2834,9 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
          * the open FD will written to the Fastflow.
          */
         case F_SET_GREEDY_MODE:
          * the open FD will written to the Fastflow.
          */
         case F_SET_GREEDY_MODE:
-               /* intentionally drop through to the same handler as F_SETSTATIC.
-                * both fcntls should pass the argument and their selector into VNOP_IOCTL.
-                */
+       /* intentionally drop through to the same handler as F_SETSTATIC.
+        * both fcntls should pass the argument and their selector into VNOP_IOCTL.
+        */
  
         /*
          * SPI (private) for indicating to a filesystem that subsequent writes to
  
         /*
          * SPI (private) for indicating to a filesystem that subsequent writes to
@@ -2359,7 +2864,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  
                 /* Only go forward if you have write access */
                 vfs_context_t ctx = vfs_context_current();
  
                 /* Only go forward if you have write access */
                 vfs_context_t ctx = vfs_context_current();
-               if(vnode_authorize(vp, NULLVP, (KAUTH_VNODE_ACCESS | KAUTH_VNODE_WRITE_DATA), ctx) != 0) {
+               if (vnode_authorize(vp, NULLVP, (KAUTH_VNODE_ACCESS | KAUTH_VNODE_WRITE_DATA), ctx) != 0) {
                         vnode_put(vp);
                         error = EBADF;
                         goto outdrop;
                         vnode_put(vp);
                         error = EBADF;
                         goto outdrop;
@@ -2384,8 +2889,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         /* extract 32 bits of flags from userland */
                         param_ptr = (caddr_t) uap->arg;
                         param = (uint32_t) param_ptr;
                         /* extract 32 bits of flags from userland */
                         param_ptr = (caddr_t) uap->arg;
                         param = (uint32_t) param_ptr;
-               }
-               else {
+               } else {
                         /* If no argument is specified, error out */
                         error = EINVAL;
                         goto out;
                         /* If no argument is specified, error out */
                         error = EINVAL;
                         goto out;
@@ -2396,12 +2900,12 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                  * all of them are mutually exclusive for now.
                  */
                 switch (param) {
                  * all of them are mutually exclusive for now.
                  */
                 switch (param) {
-                       case F_IOTYPE_ISOCHRONOUS:
-                               break;
+               case F_IOTYPE_ISOCHRONOUS:
+                       break;
  
  
-                       default:
-                               error = EINVAL;
-                               goto out;
+               default:
+                       error = EINVAL;
+                       goto out;
                 }
  
  
                 }
  
  
@@ -2420,7 +2924,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  
                 /* Only go forward if you have write access */
                 vfs_context_t ctx = vfs_context_current();
  
                 /* Only go forward if you have write access */
                 vfs_context_t ctx = vfs_context_current();
-               if(vnode_authorize(vp, NULLVP, (KAUTH_VNODE_ACCESS | KAUTH_VNODE_WRITE_DATA), ctx) != 0) {
+               if (vnode_authorize(vp, NULLVP, (KAUTH_VNODE_ACCESS | KAUTH_VNODE_WRITE_DATA), ctx) != 0) {
                         vnode_put(vp);
                         error = EBADF;
                         goto outdrop;
                         vnode_put(vp);
                         error = EBADF;
                         goto outdrop;
@@ -2458,7 +2962,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  
                 /* only proceed if you have write access */
                 vfs_context_t ctx = vfs_context_current();
  
                 /* only proceed if you have write access */
                 vfs_context_t ctx = vfs_context_current();
-               if(vnode_authorize(vp, NULLVP, (KAUTH_VNODE_ACCESS | KAUTH_VNODE_WRITE_DATA), ctx) != 0) {
+               if (vnode_authorize(vp, NULLVP, (KAUTH_VNODE_ACCESS | KAUTH_VNODE_WRITE_DATA), ctx) != 0) {
                         vnode_put(vp);
                         error = EBADF;
                         goto outdrop;
                         vnode_put(vp);
                         error = EBADF;
                         goto outdrop;
@@ -2467,10 +2971,9 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  
                 /* If arg != 0, set, otherwise unset */
                 if (uap->arg) {
  
                 /* If arg != 0, set, otherwise unset */
                 if (uap->arg) {
-                       error = VNOP_IOCTL (vp, uap->cmd, (caddr_t)1, 0, &context);
-               }
-               else {
-                       error = VNOP_IOCTL (vp, uap->cmd, (caddr_t)NULL, 0, &context);
+                       error = VNOP_IOCTL(vp, uap->cmd, (caddr_t)1, 0, &context);
+               } else {
+                       error = VNOP_IOCTL(vp, uap->cmd, (caddr_t)NULL, 0, &context);
                 }
  
                 vnode_put(vp);
                 }
  
                 vnode_put(vp);
@@ -2493,18 +2996,15 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 proc_fdunlock(p);
  
                 pathlen = MAXPATHLEN;
                 proc_fdunlock(p);
  
                 pathlen = MAXPATHLEN;
-               MALLOC(pathbufp, char *, pathlen, M_TEMP, M_WAITOK);
-               if (pathbufp == NULL) {
-                       error = ENOMEM;
-                       goto outdrop;
-               }
-               if ( (error = vnode_getwithref(vp)) == 0 ) {
+               pathbufp = zalloc(ZV_NAMEI);
+
+               if ((error = vnode_getwithref(vp)) == 0) {
                         int backingstore = 0;
  
                         /* Check for error from vn_getpath before moving on */
                         if ((error = vn_getpath(vp, pathbufp, &pathlen)) == 0) {
                                 if (vp->v_tag == VT_HFS) {
                         int backingstore = 0;
  
                         /* Check for error from vn_getpath before moving on */
                         if ((error = vn_getpath(vp, pathbufp, &pathlen)) == 0) {
                                 if (vp->v_tag == VT_HFS) {
-                                       error = VNOP_IOCTL (vp, uap->cmd, (caddr_t) &backingstore, 0, &context);
+                                       error = VNOP_IOCTL(vp, uap->cmd, (caddr_t) &backingstore, 0, &context);
                                 }
                                 (void)vnode_put(vp);
  
                                 }
                                 (void)vnode_put(vp);
  
@@ -2521,10 +3021,12 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                                                 error = EBUSY;
                                         }
                                 }
                                                 error = EBUSY;
                                         }
                                 }
-                       } else
+                       } else {
                                 (void)vnode_put(vp);
                                 (void)vnode_put(vp);
+                       }
                 }
                 }
-               FREE(pathbufp, M_TEMP);
+
+               zfree(ZV_NAMEI, pathbufp);
                 goto outdrop;
         }
  
                 goto outdrop;
         }
  
@@ -2548,24 +3050,27 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                  * for this vnode; this can include special devices, and will
                  * effectively overload fcntl() to send ioctl()'s.
                  */
                  * for this vnode; this can include special devices, and will
                  * effectively overload fcntl() to send ioctl()'s.
                  */
-               if((uap->cmd & IOC_VOID) && (uap->cmd & IOC_INOUT)){
-            error = EINVAL;
+               if ((uap->cmd & IOC_VOID) && (uap->cmd & IOC_INOUT)) {
+                       error = EINVAL;
                         goto out;
                 }
  
                 /* Catch any now-invalid fcntl() selectors */
                 switch (uap->cmd) {
                         goto out;
                 }
  
                 /* Catch any now-invalid fcntl() selectors */
                 switch (uap->cmd) {
-                       case (int)APFSIOC_REVERT_TO_SNAPSHOT:
-                       case (int)FSIOC_FIOSEEKHOLE:
-                       case (int)FSIOC_FIOSEEKDATA:
-                       case HFS_GET_BOOT_INFO:
-                       case HFS_SET_BOOT_INFO:
-                       case FIOPINSWAP:
-                       case F_MARKDEPENDENCY:
-                               error = EINVAL;
-                               goto out;
-                       default:
-                               break;
+               case (int)APFSIOC_REVERT_TO_SNAPSHOT:
+               case (int)FSIOC_FIOSEEKHOLE:
+               case (int)FSIOC_FIOSEEKDATA:
+               case (int)FSIOC_CAS_BSDFLAGS:
+               case HFS_GET_BOOT_INFO:
+               case HFS_SET_BOOT_INFO:
+               case FIOPINSWAP:
+               case F_MARKDEPENDENCY:
+               case TIOCREVOKE:
+               case TIOCREVOKECLEAR:
+                       error = EINVAL;
+                       goto out;
+               default:
+                       break;
                 }
  
                 if (fp->f_type != DTYPE_VNODE) {
                 }
  
                 if (fp->f_type != DTYPE_VNODE) {
@@ -2575,7 +3080,7 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                 vp = (struct vnode *)fp->f_data;
                 proc_fdunlock(p);
  
                 vp = (struct vnode *)fp->f_data;
                 proc_fdunlock(p);
  
-               if ( (error = vnode_getwithref(vp)) == 0 ) {
+               if ((error = vnode_getwithref(vp)) == 0) {
  #define STK_PARAMS 128
                         char stkbuf[STK_PARAMS] = {0};
                         unsigned int size;
  #define STK_PARAMS 128
                         char stkbuf[STK_PARAMS] = {0};
                         unsigned int size;
@@ -2598,8 +3103,9 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         }
  
                         memp = NULL;
                         }
  
                         memp = NULL;
-                       if (size > sizeof (stkbuf)) {
-                               if ((memp = (caddr_t)kalloc(size)) == 0) {
+                       if (size > sizeof(stkbuf)) {
+                               memp = (caddr_t)kheap_alloc(KHEAP_TEMP, size, Z_WAITOK);
+                               if (memp == 0) {
                                         (void)vnode_put(vp);
                                         error = ENOMEM;
                                         goto outdrop;
                                         (void)vnode_put(vp);
                                         error = ENOMEM;
                                         goto outdrop;
@@ -2615,14 +3121,15 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                                         error = copyin(argp, data, size);
                                         if (error) {
                                                 (void)vnode_put(vp);
                                         error = copyin(argp, data, size);
                                         if (error) {
                                                 (void)vnode_put(vp);
-                                               if (memp)
-                                                       kfree(memp, size);
+                                               if (memp) {
+                                                       kheap_free(KHEAP_TEMP, memp, size);
+                                               }
                                                 goto outdrop;
                                         }
  
                                         /* Bzero the section beyond that which was needed */
                                         if (size <= sizeof(stkbuf)) {
                                                 goto outdrop;
                                         }
  
                                         /* Bzero the section beyond that which was needed */
                                         if (size <= sizeof(stkbuf)) {
-                                               bzero ( (((uint8_t*)data) + size), (sizeof(stkbuf) - size));
+                                               bzero((((uint8_t*)data) + size), (sizeof(stkbuf) - size));
                                         }
                                 } else {
                                         /* int */
                                         }
                                 } else {
                                         /* int */
@@ -2640,9 +3147,9 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                                 bzero(data, size);
                         } else if (uap->cmd & IOC_VOID) {
                                 if (is64bit) {
                                 bzero(data, size);
                         } else if (uap->cmd & IOC_VOID) {
                                 if (is64bit) {
-                                   *(user_addr_t *)data = argp;
+                                       *(user_addr_t *)data = argp;
                                 } else {
                                 } else {
-                                   *(uint32_t *)data = (uint32_t)argp;
+                                       *(uint32_t *)data = (uint32_t)argp;
                                 }
                         }
  
                                 }
                         }
  
@@ -2651,10 +3158,12 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
                         (void)vnode_put(vp);
  
                         /* Copy any output data to user */
                         (void)vnode_put(vp);
  
                         /* Copy any output data to user */
-                       if (error == 0 && (uap->cmd & IOC_OUT) && size)
+                       if (error == 0 && (uap->cmd & IOC_OUT) && size) {
                                 error = copyout(data, argp, size);
                                 error = copyout(data, argp, size);
-                       if (memp)
-                               kfree(memp, size);
+                       }
+                       if (memp) {
+                               kheap_free(KHEAP_TEMP, memp, size);
+                       }
                 }
                 break;
         }
                 }
                 break;
         }
@@ -2662,11 +3171,11 @@ fcntl_nocancel(proc_t p, struct fcntl_nocancel_args *uap, int32_t *retval)
  outdrop:
         AUDIT_ARG(vnpath_withref, vp, ARG_VNODE1);
         fp_drop(p, fd, fp, 0);
  outdrop:
         AUDIT_ARG(vnpath_withref, vp, ARG_VNODE1);
         fp_drop(p, fd, fp, 0);
-       return(error);
+       return error;
  out:
         fp_drop(p, fd, fp, 1);
         proc_fdunlock(p);
  out:
         fp_drop(p, fd, fp, 1);
         proc_fdunlock(p);
-       return(error);
+       return error;
  }
  
  
  }
  
  
@@ -2711,16 +3220,14 @@ finishdup(proc_t p,
         if ((ofp = fdp->fd_ofiles[old]) == NULL ||
             (fdp->fd_ofileflags[old] & UF_RESERVED)) {
                 fdrelse(p, new);
         if ((ofp = fdp->fd_ofiles[old]) == NULL ||
             (fdp->fd_ofileflags[old] & UF_RESERVED)) {
                 fdrelse(p, new);
-               return (EBADF);
+               return EBADF;
         }
         }
-       fg_ref(ofp);
  
  #if CONFIG_MACF
  
  #if CONFIG_MACF
-       error = mac_file_check_dup(proc_ucred(p), ofp->f_fglob, new);
+       error = mac_file_check_dup(proc_ucred(p), ofp->fp_glob, new);
         if (error) {
         if (error) {
-               fg_drop(ofp);
                 fdrelse(p, new);
                 fdrelse(p, new);
-               return (error);
+               return error;
         }
  #endif
  
         }
  #endif
  
@@ -2731,31 +3238,34 @@ finishdup(proc_t p,
         proc_fdlock(p);
  
         if (nfp == NULL) {
         proc_fdlock(p);
  
         if (nfp == NULL) {
-               fg_drop(ofp);
                 fdrelse(p, new);
                 fdrelse(p, new);
-               return (ENOMEM);
+               return ENOMEM;
         }
  
         }
  
-       nfp->f_fglob = ofp->f_fglob;
+       fg_ref(p, ofp->fp_glob);
+       nfp->fp_glob = ofp->fp_glob;
  
  #if DIAGNOSTIC
  
  #if DIAGNOSTIC
-       if (fdp->fd_ofiles[new] != 0)
+       if (fdp->fd_ofiles[new] != 0) {
                 panic("finishdup: overwriting fd_ofiles with new %d", new);
                 panic("finishdup: overwriting fd_ofiles with new %d", new);
-       if ((fdp->fd_ofileflags[new] & UF_RESERVED) == 0)
+       }
+       if ((fdp->fd_ofileflags[new] & UF_RESERVED) == 0) {
                 panic("finishdup: unreserved fileflags with new %d", new);
                 panic("finishdup: unreserved fileflags with new %d", new);
+       }
  #endif
  
  #endif
  
-       if (new > fdp->fd_lastfile)
+       if (new > fdp->fd_lastfile) {
                 fdp->fd_lastfile = new;
                 fdp->fd_lastfile = new;
+       }
         *fdflags(p, new) |= fd_flags;
         procfdtbl_releasefd(p, new, nfp);
         *retval = new;
         *fdflags(p, new) |= fd_flags;
         procfdtbl_releasefd(p, new, nfp);
         *retval = new;
-       return (0);
+       return 0;
  }
  
  
  /*
  }
  
  
  /*
- * close
+ * sys_close
   *
   * Description:        The implementation of the close(2) system call
   *
   *
   * Description:        The implementation of the close(2) system call
   *
@@ -2768,167 +3278,141 @@ finishdup(proc_t p,
   *     fp_lookup:EBADF                 Bad file descriptor
   *      fp_guard_exception:???          Guarded file descriptor
   *     close_internal:EBADF
   *     fp_lookup:EBADF                 Bad file descriptor
   *      fp_guard_exception:???          Guarded file descriptor
   *     close_internal:EBADF
- *     close_internal:???              Anything returnable by a per-fileops
+ *     close_internal:???              Anything returnable by a per-fileops
   *                                     close function
   */
  int
   *                                     close function
   */
  int
-close(proc_t p, struct close_args *uap, int32_t *retval)
+sys_close(proc_t p, struct close_args *uap, __unused int32_t *retval)
  {
         __pthread_testcancel(1);
  {
         __pthread_testcancel(1);
-       return(close_nocancel(p, (struct close_nocancel_args *)uap, retval));
+       return close_nocancel(p, uap->fd);
  }
  
  }
  
+int
+sys_close_nocancel(proc_t p, struct close_nocancel_args *uap, __unused int32_t *retval)
+{
+       return close_nocancel(p, uap->fd);
+}
  
  int
  
  int
-close_nocancel(proc_t p, struct close_nocancel_args *uap, __unused int32_t *retval)
+close_nocancel(proc_t p, int fd)
  {
         struct fileproc *fp;
  {
         struct fileproc *fp;
-       int fd = uap->fd;
-       int error;
  
         AUDIT_SYSCLOSE(p, fd);
  
         proc_fdlock(p);
  
         AUDIT_SYSCLOSE(p, fd);
  
         proc_fdlock(p);
-
-       if ( (error = fp_lookup(p,fd,&fp, 1)) ) {
+       if ((fp = fp_get_noref_locked(p, fd)) == NULL) {
                 proc_fdunlock(p);
                 proc_fdunlock(p);
-               return(error);
+               return EBADF;
         }
  
         }
  
-       if (FP_ISGUARDED(fp, GUARD_CLOSE)) {
-               error = fp_guard_exception(p, fd, fp, kGUARD_EXC_CLOSE);
-               (void) fp_drop(p, fd, fp, 1);
+       if (fp_isguarded(fp, GUARD_CLOSE)) {
+               int error = fp_guard_exception(p, fd, fp, kGUARD_EXC_CLOSE);
                 proc_fdunlock(p);
                 proc_fdunlock(p);
-               return (error);
+               return error;
         }
  
         }
  
-       error = close_internal_locked(p, fd, fp, 0);
-
-       proc_fdunlock(p);
-
-       return (error);
+       return fp_close_and_unlock(p, fd, fp, 0);
  }
  
  
  }
  
  
-/*
- * close_internal_locked
- *
- * Close a file descriptor.
- *
- * Parameters: p                       Process in whose per process file table
- *                                     the close is to occur
- *             fd                      fd to be closed
- *             fp                      fileproc associated with the fd
- *
- * Returns:    0                       Success
- *             EBADF                   fd already in close wait state
- *     closef_locked:???               Anything returnable by a per-fileops
- *                                     close function
- *
- * Locks:      Assumes proc_fdlock for process is held by the caller and returns
- *             with lock held
- *
- * Notes:      This function may drop and reacquire this lock; it is unsafe
- *             for a caller to assume that other state protected by the lock
- *             has not been subsequently changed out from under it.
- */
  int
  int
-close_internal_locked(proc_t p, int fd, struct fileproc *fp, int flags)
+fp_close_and_unlock(proc_t p, int fd, struct fileproc *fp, int flags)
  {
         struct filedesc *fdp = p->p_fd;
  {
         struct filedesc *fdp = p->p_fd;
-       int error =0;
-       int resvfd = flags & FD_DUP2RESV;
-
+       struct fileglob *fg = fp->fp_glob;
  
  #if DIAGNOSTIC
         proc_fdlock_assert(p, LCK_MTX_ASSERT_OWNED);
  #endif
  
  
  #if DIAGNOSTIC
         proc_fdlock_assert(p, LCK_MTX_ASSERT_OWNED);
  #endif
  
-       /* Keep people from using the filedesc while we are closing it */
-       procfdtbl_markclosefd(p, fd);
-
-
-       if ((fp->f_flags & FP_CLOSING) == FP_CLOSING) {
-               panic("close_internal_locked: being called on already closing fd");
+       /*
+        * Keep most people from finding the filedesc while we are closing it.
+        *
+        * Callers are:
+        *
+        * - dup2() which always waits for UF_RESERVED to clear
+        *
+        * - close/guarded_close/... who will fail the fileproc lookup if
+        *   UF_RESERVED is set,
+        *
+        * - fdexec()/fdfree() who only run once all threads in the proc
+        *   are properly canceled, hence no fileproc in this proc should
+        *   be in flux.
+        *
+        * Which means that neither UF_RESERVED nor UF_CLOSING should be set.
+        *
+        * Callers of fp_get_noref_locked_with_iocount() can still find
+        * this entry so that they can drop their I/O reference despite
+        * not having remembered the fileproc pointer (namely select() and
+        * file_drop()).
+        */
+       if (p->p_fd->fd_ofileflags[fd] & (UF_RESERVED | UF_CLOSING)) {
+               panic("%s: called with fileproc in flux (%d/:%p)",
+                   __func__, fd, fp);
         }
         }
+       p->p_fd->fd_ofileflags[fd] |= (UF_RESERVED | UF_CLOSING);
  
  
+       if ((fp->fp_flags & FP_AIOISSUED) || kauth_authorize_fileop_has_listeners()) {
+               proc_fdunlock(p);
  
  
-#if DIAGNOSTIC
-       if ((fdp->fd_ofileflags[fd] & UF_RESERVED) == 0)
-               panic("close_internal: unreserved fileflags with fd %d", fd);
-#endif
-
-       fp->f_flags |= FP_CLOSING;
-
-       if ( (fp->f_flags & FP_AIOISSUED) || kauth_authorize_fileop_has_listeners() ) {
-
-               proc_fdunlock(p);
-
-               if ( (fp->f_type == DTYPE_VNODE) && kauth_authorize_fileop_has_listeners() ) {
-                       /*
+               if ((FILEGLOB_DTYPE(fg) == DTYPE_VNODE) && kauth_authorize_fileop_has_listeners()) {
+                       /*
                          * call out to allow 3rd party notification of close.
                          * Ignore result of kauth_authorize_fileop call.
                          */
                          * call out to allow 3rd party notification of close.
                          * Ignore result of kauth_authorize_fileop call.
                          */
-                       if (vnode_getwithref((vnode_t)fp->f_data) == 0) {
-                               u_int   fileop_flags = 0;
-                               if ((fp->f_flags & FP_WRITTEN) != 0)
-                                       fileop_flags |= KAUTH_FILEOP_CLOSE_MODIFIED;
-                               kauth_authorize_fileop(fp->f_fglob->fg_cred, KAUTH_FILEOP_CLOSE,
-                                                      (uintptr_t)fp->f_data, (uintptr_t)fileop_flags);
-                               vnode_put((vnode_t)fp->f_data);
+                       if (vnode_getwithref((vnode_t)fg->fg_data) == 0) {
+                               u_int   fileop_flags = 0;
+                               if (fg->fg_flag & FWASWRITTEN) {
+                                       fileop_flags |= KAUTH_FILEOP_CLOSE_MODIFIED;
+                               }
+                               kauth_authorize_fileop(fg->fg_cred, KAUTH_FILEOP_CLOSE,
+                                   (uintptr_t)fg->fg_data, (uintptr_t)fileop_flags);
+#if CONFIG_MACF
+                               mac_file_notify_close(proc_ucred(p), fp->fp_glob);
+#endif
+                               vnode_put((vnode_t)fg->fg_data);
                         }
                 }
                         }
                 }
-               if (fp->f_flags & FP_AIOISSUED)
-                       /*
+               if (fp->fp_flags & FP_AIOISSUED) {
+                       /*
                          * cancel all async IO requests that can be cancelled.
                          */
                          * cancel all async IO requests that can be cancelled.
                          */
-                       _aio_close( p, fd );
+                       _aio_close( p, fd );
+               }
  
                 proc_fdlock(p);
         }
  
  
                 proc_fdlock(p);
         }
  
-       if (fd < fdp->fd_knlistsize)
+       if (fd < fdp->fd_knlistsize) {
                 knote_fdclose(p, fd);
                 knote_fdclose(p, fd);
-
-       if (fp->f_flags & FP_WAITEVENT)
-               (void)waitevent_close(p, fp);
+       }
  
         fileproc_drain(p, fp);
  
  
         fileproc_drain(p, fp);
  
-       if (resvfd == 0) {
-               _fdrelse(p, fd);
+       if (flags & FD_DUP2RESV) {
+               fdp->fd_ofiles[fd] = NULL;
+               fdp->fd_ofileflags[fd] &= ~(UF_CLOSING | UF_EXCLOSE | UF_FORKCLOSE);
         } else {
         } else {
-               procfdtbl_reservefd(p, fd);
+               fdrelse(p, fd);
         }
  
         }
  
-       if (ENTR_SHOULDTRACE && fp->f_type == DTYPE_SOCKET)
-               KERNEL_ENERGYTRACE(kEnTrActKernSocket, DBG_FUNC_END,
-                   fd, 0, (int64_t)VM_KERNEL_ADDRPERM(fp->f_data));
-
-       error = closef_locked(fp, fp->f_fglob, p);
-       if ((fp->f_flags & FP_WAITCLOSE) == FP_WAITCLOSE)
-               wakeup(&fp->f_flags);
-       fp->f_flags &= ~(FP_WAITCLOSE | FP_CLOSING);
-
         proc_fdunlock(p);
  
         proc_fdunlock(p);
  
-       fileproc_free(fp);
-
-       proc_fdlock(p);
-
-#if DIAGNOSTIC
-       if (resvfd != 0) {
-               if ((fdp->fd_ofileflags[fd] & UF_RESERVED) == 0)
-                       panic("close with reserved fd returns with freed fd:%d: proc: %p", fd, p);
+       if (ENTR_SHOULDTRACE && FILEGLOB_DTYPE(fg) == DTYPE_SOCKET) {
+               KERNEL_ENERGYTRACE(kEnTrActKernSocket, DBG_FUNC_END,
+                   fd, 0, (int64_t)VM_KERNEL_ADDRPERM(fg->fg_data));
         }
         }
-#endif
  
  
-       return(error);
+       fileproc_free(fp);
+
+       return fg_drop(p, fg);
  }
  
  
  /*
  }
  
  
  /*
- * fstat1
+ * fstat
   *
   * Description:        Return status information about a file descriptor.
   *
   *
   * Description:        Return status information about a file descriptor.
   *
@@ -2962,7 +3446,7 @@ close_internal_locked(proc_t p, int fd, struct fileproc *fp, int flags)
   *             XXX fileops instead.
   */
  static int
   *             XXX fileops instead.
   */
  static int
-fstat1(proc_t p, int fd, user_addr_t ub, user_addr_t xsecurity, user_addr_t xsecurity_size, int isstat64)
+fstat(proc_t p, int fd, user_addr_t ub, user_addr_t xsecurity, user_addr_t xsecurity_size, int isstat64)
  {
         struct fileproc *fp;
         union {
  {
         struct fileproc *fp;
         union {
@@ -2987,7 +3471,7 @@ fstat1(proc_t p, int fd, user_addr_t ub, user_addr_t xsecurity, user_addr_t xsec
         AUDIT_ARG(fd, fd);
  
         if ((error = fp_lookup(p, fd, &fp, 0)) != 0) {
         AUDIT_ARG(fd, fd);
  
         if ((error = fp_lookup(p, fd, &fp, 0)) != 0) {
-               return(error);
+               return error;
         }
         type = fp->f_type;
         data = fp->f_data;
         }
         type = fp->f_type;
         data = fp->f_data;
@@ -2996,7 +3480,6 @@ fstat1(proc_t p, int fd, user_addr_t ub, user_addr_t xsecurity, user_addr_t xsec
         sbptr = (void *)&source;
  
         switch (type) {
         sbptr = (void *)&source;
  
         switch (type) {
-
         case DTYPE_VNODE:
                 if ((error = vnode_getwithref((vnode_t)data)) == 0) {
                         /*
         case DTYPE_VNODE:
                 if ((error = vnode_getwithref((vnode_t)data)) == 0) {
                         /*
@@ -3005,10 +3488,10 @@ fstat1(proc_t p, int fd, user_addr_t ub, user_addr_t xsecurity, user_addr_t xsec
                          * going to let them get the basic stat information.
                          */
                         if (xsecurity == USER_ADDR_NULL) {
                          * going to let them get the basic stat information.
                          */
                         if (xsecurity == USER_ADDR_NULL) {
-                               error = vn_stat_noauth((vnode_t)data, sbptr, NULL, isstat64, ctx,
-                                                      fp->f_fglob->fg_cred);
+                               error = vn_stat_noauth((vnode_t)data, sbptr, NULL, isstat64, 0, ctx,
+                                   fp->fp_glob->fg_cred);
                         } else {
                         } else {
-                               error = vn_stat((vnode_t)data, sbptr, &fsec, isstat64, ctx);
+                               error = vn_stat((vnode_t)data, sbptr, &fsec, isstat64, 0, ctx);
                         }
  
                         AUDIT_ARG(vnpath, (struct vnode *)data, ARG_VNODE1);
                         }
  
                         AUDIT_ARG(vnpath, (struct vnode *)data, ARG_VNODE1);
@@ -3075,9 +3558,8 @@ fstat1(proc_t p, int fd, user_addr_t ub, user_addr_t xsecurity, user_addr_t xsec
  
         /* caller wants extended security information? */
         if (xsecurity != USER_ADDR_NULL) {
  
         /* caller wants extended security information? */
         if (xsecurity != USER_ADDR_NULL) {
-
                 /* did we get any? */
                 /* did we get any? */
-                if (fsec == KAUTH_FILESEC_NONE) {
+               if (fsec == KAUTH_FILESEC_NONE) {
                         if (susize(xsecurity_size, 0) != 0) {
                                 error = EFAULT;
                                 goto out;
                         if (susize(xsecurity_size, 0) != 0) {
                                 error = EFAULT;
                                 goto out;
@@ -3093,20 +3575,22 @@ fstat1(proc_t p, int fd, user_addr_t ub, user_addr_t xsecurity, user_addr_t xsec
                         }
  
                         /* if the caller supplied enough room, copy out to it */
                         }
  
                         /* if the caller supplied enough room, copy out to it */
-                       if (xsecurity_bufsize >= KAUTH_FILESEC_COPYSIZE(fsec))
+                       if (xsecurity_bufsize >= KAUTH_FILESEC_COPYSIZE(fsec)) {
                                 error = copyout(fsec, xsecurity, KAUTH_FILESEC_COPYSIZE(fsec));
                                 error = copyout(fsec, xsecurity, KAUTH_FILESEC_COPYSIZE(fsec));
+                       }
                 }
         }
  out:
         fp_drop(p, fd, fp, 0);
                 }
         }
  out:
         fp_drop(p, fd, fp, 0);
-       if (fsec != NULL)
+       if (fsec != NULL) {
                 kauth_filesec_free(fsec);
                 kauth_filesec_free(fsec);
-       return (error);
+       }
+       return error;
  }
  
  
  /*
  }
  
  
  /*
- * fstat_extended
+ * sys_fstat_extended
   *
   * Description:        Extended version of fstat supporting returning extended
   *             security information
   *
   * Description:        Extended version of fstat supporting returning extended
   *             security information
@@ -3119,17 +3603,17 @@ out:
   *             uap->xsecurity_size             The size of xsecurity, or 0
   *
   * Returns:    0                               Success
   *             uap->xsecurity_size             The size of xsecurity, or 0
   *
   * Returns:    0                               Success
- *             !0                              Errno (see fstat1)
+ *             !0                              Errno (see fstat)
   */
  int
   */
  int
-fstat_extended(proc_t p, struct fstat_extended_args *uap, __unused int32_t *retval)
+sys_fstat_extended(proc_t p, struct fstat_extended_args *uap, __unused int32_t *retval)
  {
  {
-       return(fstat1(p, uap->fd, uap->ub, uap->xsecurity, uap->xsecurity_size, 0));
+       return fstat(p, uap->fd, uap->ub, uap->xsecurity, uap->xsecurity_size, 0);
  }
  
  
  /*
  }
  
  
  /*
- * fstat
+ * sys_fstat
   *
   * Description:        Get file status for the file associated with fd
   *
   *
   * Description:        Get file status for the file associated with fd
   *
@@ -3138,17 +3622,17 @@ fstat_extended(proc_t p, struct fstat_extended_args *uap, __unused int32_t *retv
   *             uap->ub                         The user stat buffer
   *
   * Returns:    0                               Success
   *             uap->ub                         The user stat buffer
   *
   * Returns:    0                               Success
- *             !0                              Errno (see fstat1)
+ *             !0                              Errno (see fstat)
   */
  int
   */
  int
-fstat(proc_t p, struct fstat_args *uap, __unused int32_t *retval)
+sys_fstat(proc_t p, struct fstat_args *uap, __unused int32_t *retval)
  {
  {
-       return(fstat1(p, uap->fd, uap->ub, 0, 0, 0));
+       return fstat(p, uap->fd, uap->ub, 0, 0, 0);
  }
  
  
  /*
  }
  
  
  /*
- * fstat64_extended
+ * sys_fstat64_extended
   *
   * Description:        Extended version of fstat64 supporting returning extended
   *             security information
   *
   * Description:        Extended version of fstat64 supporting returning extended
   *             security information
@@ -3161,17 +3645,17 @@ fstat(proc_t p, struct fstat_args *uap, __unused int32_t *retval)
   *             uap->xsecurity_size             The size of xsecurity, or 0
   *
   * Returns:    0                               Success
   *             uap->xsecurity_size             The size of xsecurity, or 0
   *
   * Returns:    0                               Success
- *             !0                              Errno (see fstat1)
+ *             !0                              Errno (see fstat)
   */
  int
   */
  int
-fstat64_extended(proc_t p, struct fstat64_extended_args *uap, __unused int32_t *retval)
+sys_fstat64_extended(proc_t p, struct fstat64_extended_args *uap, __unused int32_t *retval)
  {
  {
-       return(fstat1(p, uap->fd, uap->ub, uap->xsecurity, uap->xsecurity_size, 1));
+       return fstat(p, uap->fd, uap->ub, uap->xsecurity, uap->xsecurity_size, 1);
  }
  
  
  /*
  }
  
  
  /*
- * fstat64
+ * sys_fstat64
   *
   * Description:        Get 64 bit version of the file status for the file associated
   *             with fd
   *
   * Description:        Get 64 bit version of the file status for the file associated
   *             with fd
@@ -3181,17 +3665,17 @@ fstat64_extended(proc_t p, struct fstat64_extended_args *uap, __unused int32_t *
   *             uap->ub                         The user stat buffer
   *
   * Returns:    0                               Success
   *             uap->ub                         The user stat buffer
   *
   * Returns:    0                               Success
- *             !0                              Errno (see fstat1)
+ *             !0                              Errno (see fstat)
   */
  int
   */
  int
-fstat64(proc_t p, struct fstat64_args *uap, __unused int32_t *retval)
+sys_fstat64(proc_t p, struct fstat64_args *uap, __unused int32_t *retval)
  {
  {
-       return(fstat1(p, uap->fd, uap->ub, 0, 0, 1));
+       return fstat(p, uap->fd, uap->ub, 0, 0, 1);
  }
  
  
  /*
  }
  
  
  /*
- * fpathconf
+ * sys_fpathconf
   *
   * Description:        Return pathconf information about a file descriptor.
   *
   *
   * Description:        Return pathconf information about a file descriptor.
   *
@@ -3210,7 +3694,7 @@ fstat64(proc_t p, struct fstat64_args *uap, __unused int32_t *retval)
   *             *retval (modified)              Returned information (numeric)
   */
  int
   *             *retval (modified)              Returned information (numeric)
   */
  int
-fpathconf(proc_t p, struct fpathconf_args *uap, int32_t *retval)
+sys_fpathconf(proc_t p, struct fpathconf_args *uap, int32_t *retval)
  {
         int fd = uap->fd;
         struct fileproc *fp;
  {
         int fd = uap->fd;
         struct fileproc *fp;
@@ -3221,13 +3705,13 @@ fpathconf(proc_t p, struct fpathconf_args *uap, int32_t *retval)
  
  
         AUDIT_ARG(fd, uap->fd);
  
  
         AUDIT_ARG(fd, uap->fd);
-       if ( (error = fp_lookup(p, fd, &fp, 0)) )
-               return(error);
+       if ((error = fp_lookup(p, fd, &fp, 0))) {
+               return error;
+       }
         type = fp->f_type;
         data = fp->f_data;
  
         switch (type) {
         type = fp->f_type;
         data = fp->f_data;
  
         switch (type) {
-
         case DTYPE_SOCKET:
                 if (uap->name != _PC_PIPE_BUF) {
                         error = EINVAL;
         case DTYPE_SOCKET:
                 if (uap->name != _PC_PIPE_BUF) {
                         error = EINVAL;
@@ -3249,8 +3733,8 @@ fpathconf(proc_t p, struct fpathconf_args *uap, int32_t *retval)
         case DTYPE_VNODE:
                 vp = (struct vnode *)data;
  
         case DTYPE_VNODE:
                 vp = (struct vnode *)data;
  
-               if ( (error = vnode_getwithref(vp)) == 0) {
-                       AUDIT_ARG(vnpath, vp, ARG_VNODE1);
+               if ((error = vnode_getwithref(vp)) == 0) {
+                       AUDIT_ARG(vnpath, vp, ARG_VNODE1);
  
                         error = vn_pathconf(vp, uap->name, retval, vfs_context_current());
  
  
                         error = vn_pathconf(vp, uap->name, retval, vfs_context_current());
  
@@ -3261,12 +3745,11 @@ fpathconf(proc_t p, struct fpathconf_args *uap, int32_t *retval)
         default:
                 error = EINVAL;
                 goto out;
         default:
                 error = EINVAL;
                 goto out;
-
         }
         /*NOTREACHED*/
  out:
         fp_drop(p, fd, fp, 0);
         }
         /*NOTREACHED*/
  out:
         fp_drop(p, fd, fp, 0);
-       return(error);
+       return error;
  }
  
  /*
  }
  
  /*
@@ -3299,9 +3782,13 @@ fdalloc(proc_t p, int want, int *result)
  {
         struct filedesc *fdp = p->p_fd;
         int i;
  {
         struct filedesc *fdp = p->p_fd;
         int i;
-       int lim, last, numfiles, oldnfiles;
+       int last, numfiles, oldnfiles;
         struct fileproc **newofiles, **ofiles;
         char *newofileflags;
         struct fileproc **newofiles, **ofiles;
         char *newofileflags;
+       rlim_t lim;
+       rlim_t nofile = proc_limitgetcur(p, RLIMIT_NOFILE, TRUE);
+
+       nofile = MIN(nofile, INT_MAX);
  
         /*
          * Search for a free descriptor starting at the higher
  
         /*
          * Search for a free descriptor starting at the higher
@@ -3312,44 +3799,50 @@ fdalloc(proc_t p, int want, int *result)
         proc_fdlock_assert(p, LCK_MTX_ASSERT_OWNED);
  #endif
  
         proc_fdlock_assert(p, LCK_MTX_ASSERT_OWNED);
  #endif
  
-       lim = min((int)p->p_rlimit[RLIMIT_NOFILE].rlim_cur, maxfiles);
+       lim = MIN(nofile, maxfilesperproc);
         for (;;) {
         for (;;) {
-               last = min(fdp->fd_nfiles, lim);
-               if ((i = want) < fdp->fd_freefile)
+               last = (int)MIN((unsigned int)fdp->fd_nfiles, (unsigned int)lim);
+               if ((i = want) < fdp->fd_freefile) {
                         i = fdp->fd_freefile;
                         i = fdp->fd_freefile;
+               }
                 for (; i < last; i++) {
                         if (fdp->fd_ofiles[i] == NULL && !(fdp->fd_ofileflags[i] & UF_RESERVED)) {
                                 procfdtbl_reservefd(p, i);
                 for (; i < last; i++) {
                         if (fdp->fd_ofiles[i] == NULL && !(fdp->fd_ofileflags[i] & UF_RESERVED)) {
                                 procfdtbl_reservefd(p, i);
-                               if (i > fdp->fd_lastfile)
+                               if (i > fdp->fd_lastfile) {
                                         fdp->fd_lastfile = i;
                                         fdp->fd_lastfile = i;
-                               if (want <= fdp->fd_freefile)
+                               }
+                               if (want <= fdp->fd_freefile) {
                                         fdp->fd_freefile = i;
                                         fdp->fd_freefile = i;
+                               }
                                 *result = i;
                                 *result = i;
-                               return (0);
+                               return 0;
                         }
                 }
  
                 /*
                  * No space in current array.  Expand?
                  */
                         }
                 }
  
                 /*
                  * No space in current array.  Expand?
                  */
-               if (fdp->fd_nfiles >= lim)
-                       return (EMFILE);
-               if (fdp->fd_nfiles < NDEXTENT)
+               if ((rlim_t)fdp->fd_nfiles >= lim) {
+                       return EMFILE;
+               }
+               if (fdp->fd_nfiles < NDEXTENT) {
                         numfiles = NDEXTENT;
                         numfiles = NDEXTENT;
-               else
+               } else {
                         numfiles = 2 * fdp->fd_nfiles;
                         numfiles = 2 * fdp->fd_nfiles;
+               }
                 /* Enforce lim */
                 /* Enforce lim */
-               if (numfiles > lim)
-                       numfiles = lim;
+               if ((rlim_t)numfiles > lim) {
+                       numfiles = (int)lim;
+               }
                 proc_fdunlock(p);
                 proc_fdunlock(p);
-               MALLOC_ZONE(newofiles, struct fileproc **,
-                               numfiles * OFILESIZE, M_OFILETABL, M_WAITOK);
+               newofiles = kheap_alloc(KM_OFILETABL, numfiles * OFILESIZE,
+                   Z_WAITOK);
                 proc_fdlock(p);
                 if (newofiles == NULL) {
                 proc_fdlock(p);
                 if (newofiles == NULL) {
-                       return (ENOMEM);
+                       return ENOMEM;
                 }
                 if (fdp->fd_nfiles >= numfiles) {
                 }
                 if (fdp->fd_nfiles >= numfiles) {
-                       FREE_ZONE(newofiles, numfiles * OFILESIZE, M_OFILETABL);
+                       kheap_free(KM_OFILETABL, newofiles, numfiles * OFILESIZE);
                         continue;
                 }
                 newofileflags = (char *) &newofiles[numfiles];
                         continue;
                 }
                 newofileflags = (char *) &newofiles[numfiles];
@@ -3359,20 +3852,20 @@ fdalloc(proc_t p, int want, int *result)
                  */
                 oldnfiles = fdp->fd_nfiles;
                 (void) memcpy(newofiles, fdp->fd_ofiles,
                  */
                 oldnfiles = fdp->fd_nfiles;
                 (void) memcpy(newofiles, fdp->fd_ofiles,
-                               oldnfiles * sizeof(*fdp->fd_ofiles));
+                   oldnfiles * sizeof(*fdp->fd_ofiles));
                 (void) memset(&newofiles[oldnfiles], 0,
                 (void) memset(&newofiles[oldnfiles], 0,
-                               (numfiles - oldnfiles) * sizeof(*fdp->fd_ofiles));
+                   (numfiles - oldnfiles) * sizeof(*fdp->fd_ofiles));
  
                 (void) memcpy(newofileflags, fdp->fd_ofileflags,
  
                 (void) memcpy(newofileflags, fdp->fd_ofileflags,
-                               oldnfiles * sizeof(*fdp->fd_ofileflags));
+                   oldnfiles * sizeof(*fdp->fd_ofileflags));
                 (void) memset(&newofileflags[oldnfiles], 0,
                 (void) memset(&newofileflags[oldnfiles], 0,
-                               (numfiles - oldnfiles) *
-                                               sizeof(*fdp->fd_ofileflags));
+                   (numfiles - oldnfiles) *
+                   sizeof(*fdp->fd_ofileflags));
                 ofiles = fdp->fd_ofiles;
                 fdp->fd_ofiles = newofiles;
                 fdp->fd_ofileflags = newofileflags;
                 fdp->fd_nfiles = numfiles;
                 ofiles = fdp->fd_ofiles;
                 fdp->fd_ofiles = newofiles;
                 fdp->fd_ofileflags = newofileflags;
                 fdp->fd_nfiles = numfiles;
-               FREE_ZONE(ofiles, oldnfiles * OFILESIZE, M_OFILETABL);
+               kheap_free(KM_OFILETABL, ofiles, oldnfiles * OFILESIZE);
                 fdexpand++;
         }
  }
                 fdexpand++;
         }
  }
@@ -3401,75 +3894,81 @@ fdavail(proc_t p, int n)
         struct filedesc *fdp = p->p_fd;
         struct fileproc **fpp;
         char *flags;
         struct filedesc *fdp = p->p_fd;
         struct fileproc **fpp;
         char *flags;
-       int i, lim;
+       int i;
+       int lim;
+       rlim_t nofile = proc_limitgetcur(p, RLIMIT_NOFILE, TRUE);
  
  
-       lim = min((int)p->p_rlimit[RLIMIT_NOFILE].rlim_cur, maxfiles);
-       if ((i = lim - fdp->fd_nfiles) > 0 && (n -= i) <= 0)
-               return (1);
+       lim = (int)MIN(nofile, maxfilesperproc);
+       if ((i = lim - fdp->fd_nfiles) > 0 && (n -= i) <= 0) {
+               return 1;
+       }
         fpp = &fdp->fd_ofiles[fdp->fd_freefile];
         flags = &fdp->fd_ofileflags[fdp->fd_freefile];
         fpp = &fdp->fd_ofiles[fdp->fd_freefile];
         flags = &fdp->fd_ofileflags[fdp->fd_freefile];
-       for (i = fdp->fd_nfiles - fdp->fd_freefile; --i >= 0; fpp++, flags++)
-               if (*fpp == NULL && !(*flags & UF_RESERVED) && --n <= 0)
-                       return (1);
-       return (0);
+       for (i = fdp->fd_nfiles - fdp->fd_freefile; --i >= 0; fpp++, flags++) {
+               if (*fpp == NULL && !(*flags & UF_RESERVED) && --n <= 0) {
+                       return 1;
+               }
+       }
+       return 0;
  }
  
  
  }
  
  
-/*
- * fdrelse
- *
- * Description:        Legacy KPI wrapper function for _fdrelse
- *
- * Parameters: p                               Process in which fd lives
- *             fd                              fd to free
- *
- * Returns:    void
- *
- * Locks:      Assumes proc_fdlock for process is held by the caller
- */
-void
-fdrelse(proc_t p, int fd)
+struct fileproc *
+fp_get_noref_locked(proc_t p, int fd)
  {
  {
-       _fdrelse(p, fd);
+       struct filedesc *fdp = p->p_fd;
+       struct fileproc *fp;
+
+       if (fd < 0 || fd >= fdp->fd_nfiles ||
+           (fp = fdp->fd_ofiles[fd]) == NULL ||
+           (fdp->fd_ofileflags[fd] & UF_RESERVED)) {
+               return NULL;
+       }
+       return fp;
  }
  
  }
  
+struct fileproc *
+fp_get_noref_locked_with_iocount(proc_t p, int fd)
+{
+       struct filedesc *fdp = p->p_fd;
+       struct fileproc *fp = NULL;
+
+       if (fd < 0 || fd >= fdp->fd_nfiles ||
+           (fp = fdp->fd_ofiles[fd]) == NULL ||
+           os_ref_get_count(&fp->fp_iocount) <= 1 ||
+           ((fdp->fd_ofileflags[fd] & UF_RESERVED) &&
+           !(fdp->fd_ofileflags[fd] & UF_CLOSING))) {
+               panic("%s: caller without an ioccount on fileproc (%d/:%p)",
+                   __func__, fd, fp);
+       }
+
+       return fp;
+}
  
  
-/*
- * fdgetf_noref
- *
- * Description:        Get the fileproc pointer for the given fd from the per process
- *             open file table without taking an explicit reference on it.
- *
- * Parameters: p                               Process containing fd
- *             fd                              fd to obtain fileproc for
- *             resultfp                        Pointer to pointer return area
- *
- * Returns:    0                               Success
- *             EBADF
- *
- * Implicit returns:
- *             *resultfp (modified)            Pointer to fileproc pointer
- *
- * Locks:      Assumes proc_fdlock for process is held by the caller
- *
- * Notes:      Because there is no reference explicitly taken, the returned
- *             fileproc pointer is only valid so long as the proc_fdlock
- *             remains held by the caller.
- */
  int
  int
-fdgetf_noref(proc_t p, int fd, struct fileproc **resultfp)
+fp_get_ftype(proc_t p, int fd, file_type_t ftype, int err, struct fileproc **fpp)
  {
         struct filedesc *fdp = p->p_fd;
         struct fileproc *fp;
  
  {
         struct filedesc *fdp = p->p_fd;
         struct fileproc *fp;
  
+       proc_fdlock_spin(p);
         if (fd < 0 || fd >= fdp->fd_nfiles ||
         if (fd < 0 || fd >= fdp->fd_nfiles ||
-                       (fp = fdp->fd_ofiles[fd]) == NULL ||
-                       (fdp->fd_ofileflags[fd] & UF_RESERVED)) {
-               return (EBADF);
+           (fp = fdp->fd_ofiles[fd]) == NULL ||
+           (fdp->fd_ofileflags[fd] & UF_RESERVED)) {
+               proc_fdunlock(p);
+               return EBADF;
         }
         }
-       if (resultfp)
-               *resultfp = fp;
-       return (0);
+
+       if (fp->f_type != ftype) {
+               proc_fdunlock(p);
+               return err;
+       }
+
+       os_ref_retain_locked(&fp->fp_iocount);
+       proc_fdunlock(p);
+
+       *fpp = fp;
+       return 0;
  }
  
  
  }
  
  
@@ -3478,7 +3977,7 @@ fdgetf_noref(proc_t p, int fd, struct fileproc **resultfp)
   *
   * Description:        Get fileproc and vnode pointer for a given fd from the per
   *             process open file table of the specified process, and if
   *
   * Description:        Get fileproc and vnode pointer for a given fd from the per
   *             process open file table of the specified process, and if
- *             successful, increment the f_iocount
+ *             successful, increment the fp_iocount
   *
   * Parameters: p                               Process in which fd lives
   *             fd                              fd to get information for
   *
   * Parameters: p                               Process in which fd lives
   *             fd                              fd to get information for
@@ -3503,438 +4002,134 @@ fdgetf_noref(proc_t p, int fd, struct fileproc **resultfp)
  int
  fp_getfvp(proc_t p, int fd, struct fileproc **resultfp, struct vnode **resultvp)
  {
  int
  fp_getfvp(proc_t p, int fd, struct fileproc **resultfp, struct vnode **resultvp)
  {
-       struct filedesc *fdp = p->p_fd;
         struct fileproc *fp;
         struct fileproc *fp;
+       int error;
  
  
-       proc_fdlock_spin(p);
-       if (fd < 0 || fd >= fdp->fd_nfiles ||
-                       (fp = fdp->fd_ofiles[fd]) == NULL ||
-                       (fdp->fd_ofileflags[fd] & UF_RESERVED)) {
-               proc_fdunlock(p);
-               return (EBADF);
-       }
-       if (fp->f_type != DTYPE_VNODE) {
-               proc_fdunlock(p);
-               return(ENOTSUP);
+       error = fp_get_ftype(p, fd, DTYPE_VNODE, ENOTSUP, &fp);
+       if (error == 0) {
+               if (resultfp) {
+                       *resultfp = fp;
+               }
+               if (resultvp) {
+                       *resultvp = (struct vnode *)fp->f_data;
+               }
         }
         }
-       fp->f_iocount++;
  
  
-       if (resultfp)
-               *resultfp = fp;
-       if (resultvp)
-               *resultvp = (struct vnode *)fp->f_data;
-       proc_fdunlock(p);
-
-       return (0);
+       return error;
  }
  
  
  /*
  }
  
  
  /*
- * fp_getfvpandvid
+ * fp_get_pipe_id
   *
   *
- * Description:        Get fileproc, vnode pointer, and vid for a given fd from the
- *             per process open file table of the specified process, and if
- *             successful, increment the f_iocount
+ * Description:        Get pipe id for a given fd from the per process open file table
+ *             of the specified process.
   *
   * Parameters: p                               Process in which fd lives
   *             fd                              fd to get information for
   *
   * Parameters: p                               Process in which fd lives
   *             fd                              fd to get information for
- *             resultfp                        Pointer to result fileproc
- *                                             pointer area, or 0 if none
- *             resultvp                        Pointer to result vnode pointer
- *                                             area, or 0 if none
- *             vidp                            Pointer to resuld vid area
+ *             result_pipe_id                  Pointer to result pipe id
   *
   * Returns:    0                               Success
   *
   * Returns:    0                               Success
- *             EBADF                           Bad file descriptor
- *             ENOTSUP                         fd does not refer to a vnode
+ *             EIVAL                           NULL pointer arguments passed
+ *             fp_lookup:EBADF                 Bad file descriptor
+ *             ENOTSUP                         fd does not refer to a pipe
   *
   * Implicit returns:
   *
   * Implicit returns:
- *             *resultfp (modified)            Fileproc pointer
- *             *resultvp (modified)            vnode pointer
- *             *vidp                           vid value
- *
- * Notes:      The resultfp and resultvp fields are optional, and may be
- *             independently specified as NULL to skip returning information
+ *             *result_pipe_id (modified)      pipe id
   *
   * Locks:      Internally takes and releases proc_fdlock
   */
  int
   *
   * Locks:      Internally takes and releases proc_fdlock
   */
  int
-fp_getfvpandvid(proc_t p, int fd, struct fileproc **resultfp,
-               struct vnode **resultvp, uint32_t *vidp)
+fp_get_pipe_id(proc_t p, int fd, uint64_t *result_pipe_id)
  {
  {
-       struct filedesc *fdp = p->p_fd;
-       struct fileproc *fp;
+       struct fileproc *fp = FILEPROC_NULL;
+       struct fileglob *fg = NULL;
+       int error = 0;
  
  
-       proc_fdlock_spin(p);
-       if (fd < 0 || fd >= fdp->fd_nfiles ||
-                       (fp = fdp->fd_ofiles[fd]) == NULL ||
-                       (fdp->fd_ofileflags[fd] & UF_RESERVED)) {
-               proc_fdunlock(p);
-               return (EBADF);
+       if (p == NULL || result_pipe_id == NULL) {
+               return EINVAL;
         }
         }
-       if (fp->f_type != DTYPE_VNODE) {
+
+       proc_fdlock(p);
+       if ((error = fp_lookup(p, fd, &fp, 1))) {
                 proc_fdunlock(p);
                 proc_fdunlock(p);
-               return(ENOTSUP);
+               return error;
         }
         }
-       fp->f_iocount++;
+       fg = fp->fp_glob;
  
  
-       if (resultfp)
-               *resultfp = fp;
-       if (resultvp)
-               *resultvp = (struct vnode *)fp->f_data;
-       if (vidp)
-               *vidp = (uint32_t)vnode_vid((struct vnode *)fp->f_data);
-       proc_fdunlock(p);
+       if (FILEGLOB_DTYPE(fg) == DTYPE_PIPE) {
+               *result_pipe_id = pipe_id((struct pipe*)fg->fg_data);
+       } else {
+               error = ENOTSUP;
+       }
  
  
-       return (0);
+       fp_drop(p, fd, fp, 1);
+       proc_fdunlock(p);
+       return error;
  }
  
  
  /*
  }
  
  
  /*
- * fp_getfsock
+ * fp_lookup
   *
   *
- * Description:        Get fileproc and socket pointer for a given fd from the
- *             per process open file table of the specified process, and if
- *             successful, increment the f_iocount
+ * Description:        Get fileproc pointer for a given fd from the per process
+ *             open file table of the specified process and if successful,
+ *             increment the fp_iocount
   *
   * Parameters: p                               Process in which fd lives
   *             fd                              fd to get information for
   *             resultfp                        Pointer to result fileproc
   *                                             pointer area, or 0 if none
   *
   * Parameters: p                               Process in which fd lives
   *             fd                              fd to get information for
   *             resultfp                        Pointer to result fileproc
   *                                             pointer area, or 0 if none
- *             results                         Pointer to result socket
- *                                             pointer area, or 0 if none
+ *             locked                          !0 if the caller holds the
+ *                                             proc_fdlock, 0 otherwise
   *
   *
- * Returns:    EBADF                   The file descriptor is invalid
- *             EOPNOTSUPP              The file descriptor is not a socket
- *             0                       Success
+ * Returns:    0                       Success
+ *             EBADF                   Bad file descriptor
   *
   * Implicit returns:
   *             *resultfp (modified)            Fileproc pointer
   *
   * Implicit returns:
   *             *resultfp (modified)            Fileproc pointer
- *             *results (modified)             socket pointer
   *
   *
- * Notes:      EOPNOTSUPP should probably be ENOTSOCK; this function is only
- *             ever called from accept1().
+ * Locks:      If the argument 'locked' is non-zero, then the caller is
+ *             expected to have taken and held the proc_fdlock; if it is
+ *             zero, than this routine internally takes and drops this lock.
   */
  int
   */
  int
-fp_getfsock(proc_t p, int fd, struct fileproc **resultfp,
-           struct socket **results)
+fp_lookup(proc_t p, int fd, struct fileproc **resultfp, int locked)
  {
         struct filedesc *fdp = p->p_fd;
         struct fileproc *fp;
  
  {
         struct filedesc *fdp = p->p_fd;
         struct fileproc *fp;
  
-       proc_fdlock_spin(p);
-       if (fd < 0 || fd >= fdp->fd_nfiles ||
-                       (fp = fdp->fd_ofiles[fd]) == NULL ||
-                       (fdp->fd_ofileflags[fd] & UF_RESERVED)) {
-               proc_fdunlock(p);
-               return (EBADF);
+       if (!locked) {
+               proc_fdlock_spin(p);
         }
         }
-       if (fp->f_type != DTYPE_SOCKET) {
-               proc_fdunlock(p);
-               return(EOPNOTSUPP);
+       if (fd < 0 || fdp == NULL || fd >= fdp->fd_nfiles ||
+           (fp = fdp->fd_ofiles[fd]) == NULL ||
+           (fdp->fd_ofileflags[fd] & UF_RESERVED)) {
+               if (!locked) {
+                       proc_fdunlock(p);
+               }
+               return EBADF;
         }
         }
-       fp->f_iocount++;
+       os_ref_retain_locked(&fp->fp_iocount);
  
  
-       if (resultfp)
+       if (resultfp) {
                 *resultfp = fp;
                 *resultfp = fp;
-       if (results)
-               *results = (struct socket *)fp->f_data;
-       proc_fdunlock(p);
+       }
+       if (!locked) {
+               proc_fdunlock(p);
+       }
  
  
-       return (0);
+       return 0;
  }
  
  
  /*
  }
  
  
  /*
- * fp_getfkq
- *
- * Description:        Get fileproc and kqueue pointer for a given fd from the
- *             per process open file table of the specified process, and if
- *             successful, increment the f_iocount
- *
- * Parameters: p                               Process in which fd lives
- *             fd                              fd to get information for
- *             resultfp                        Pointer to result fileproc
- *                                             pointer area, or 0 if none
- *             resultkq                        Pointer to result kqueue
- *                                             pointer area, or 0 if none
- *
- * Returns:    EBADF                   The file descriptor is invalid
- *             EBADF                   The file descriptor is not a socket
- *             0                       Success
- *
- * Implicit returns:
- *             *resultfp (modified)            Fileproc pointer
- *             *resultkq (modified)            kqueue pointer
- *
- * Notes:      The second EBADF should probably be something else to make
- *             the error condition distinct.
- */
-int
-fp_getfkq(proc_t p, int fd, struct fileproc **resultfp,
-         struct kqueue **resultkq)
-{
-       struct filedesc *fdp = p->p_fd;
-       struct fileproc *fp;
-
-       proc_fdlock_spin(p);
-       if ( fd < 0 || fd >= fdp->fd_nfiles ||
-                       (fp = fdp->fd_ofiles[fd]) == NULL ||
-                       (fdp->fd_ofileflags[fd] & UF_RESERVED)) {
-               proc_fdunlock(p);
-               return (EBADF);
-       }
-       if (fp->f_type != DTYPE_KQUEUE) {
-               proc_fdunlock(p);
-               return(EBADF);
-       }
-       fp->f_iocount++;
-
-       if (resultfp)
-               *resultfp = fp;
-       if (resultkq)
-               *resultkq = (struct kqueue *)fp->f_data;
-       proc_fdunlock(p);
-
-       return (0);
-}
-
-
-/*
- * fp_getfpshm
- *
- * Description:        Get fileproc and POSIX shared memory pointer for a given fd
- *             from the per process open file table of the specified process
- *             and if successful, increment the f_iocount
- *
- * Parameters: p                               Process in which fd lives
- *             fd                              fd to get information for
- *             resultfp                        Pointer to result fileproc
- *                                             pointer area, or 0 if none
- *             resultpshm                      Pointer to result POSIX
- *                                             shared memory pointer
- *                                             pointer area, or 0 if none
- *
- * Returns:    EBADF                   The file descriptor is invalid
- *             EBADF                   The file descriptor is not a POSIX
- *                                     shared memory area
- *             0                       Success
- *
- * Implicit returns:
- *             *resultfp (modified)            Fileproc pointer
- *             *resultpshm (modified)          POSIX shared memory pointer
- *
- * Notes:      The second EBADF should probably be something else to make
- *             the error condition distinct.
- */
-int
-fp_getfpshm(proc_t p, int fd, struct fileproc **resultfp,
-           struct pshmnode **resultpshm)
-{
-       struct filedesc *fdp = p->p_fd;
-       struct fileproc *fp;
-
-       proc_fdlock_spin(p);
-       if (fd < 0 || fd >= fdp->fd_nfiles ||
-                       (fp = fdp->fd_ofiles[fd]) == NULL ||
-                       (fdp->fd_ofileflags[fd] & UF_RESERVED)) {
-               proc_fdunlock(p);
-               return (EBADF);
-       }
-       if (fp->f_type != DTYPE_PSXSHM) {
-
-               proc_fdunlock(p);
-               return(EBADF);
-       }
-       fp->f_iocount++;
-
-       if (resultfp)
-               *resultfp = fp;
-       if (resultpshm)
-               *resultpshm = (struct pshmnode *)fp->f_data;
-       proc_fdunlock(p);
-
-       return (0);
-}
-
-
-/*
- * fp_getfsem
- *
- * Description:        Get fileproc and POSIX semaphore pointer for a given fd from
- *             the per process open file table of the specified process
- *             and if successful, increment the f_iocount
- *
- * Parameters: p                               Process in which fd lives
- *             fd                              fd to get information for
- *             resultfp                        Pointer to result fileproc
- *                                             pointer area, or 0 if none
- *             resultpsem                      Pointer to result POSIX
- *                                             semaphore pointer area, or
- *                                             0 if none
- *
- * Returns:    EBADF                   The file descriptor is invalid
- *             EBADF                   The file descriptor is not a POSIX
- *                                     semaphore
- *             0                       Success
- *
- * Implicit returns:
- *             *resultfp (modified)            Fileproc pointer
- *             *resultpsem (modified)          POSIX semaphore pointer
- *
- * Notes:      The second EBADF should probably be something else to make
- *             the error condition distinct.
- *
- *             In order to support unnamed POSIX semaphores, the named
- *             POSIX semaphores will have to move out of the per-process
- *             open filetable, and into a global table that is shared with
- *             unnamed POSIX semaphores, since unnamed POSIX semaphores
- *             are typically used by declaring instances in shared memory,
- *             and there's no other way to do this without changing the
- *             underlying type, which would introduce binary compatibility
- *             issues.
- */
-int
-fp_getfpsem(proc_t p, int fd, struct fileproc **resultfp,
-           struct psemnode **resultpsem)
-{
-       struct filedesc *fdp = p->p_fd;
-       struct fileproc *fp;
-
-       proc_fdlock_spin(p);
-       if (fd < 0 || fd >= fdp->fd_nfiles ||
-                       (fp = fdp->fd_ofiles[fd]) == NULL ||
-                       (fdp->fd_ofileflags[fd] & UF_RESERVED)) {
-               proc_fdunlock(p);
-               return (EBADF);
-       }
-       if (fp->f_type != DTYPE_PSXSEM) {
-               proc_fdunlock(p);
-               return(EBADF);
-       }
-       fp->f_iocount++;
-
-       if (resultfp)
-               *resultfp = fp;
-       if (resultpsem)
-               *resultpsem = (struct psemnode *)fp->f_data;
-       proc_fdunlock(p);
-
-       return (0);
-}
-
-
-/*
- * fp_getfpipe
- *
- * Description:        Get fileproc and pipe pointer for a given fd from the
- *             per process open file table of the specified process
- *             and if successful, increment the f_iocount
- *
- * Parameters: p                               Process in which fd lives
- *             fd                              fd to get information for
- *             resultfp                        Pointer to result fileproc
- *                                             pointer area, or 0 if none
- *             resultpipe                      Pointer to result pipe
- *                                             pointer area, or 0 if none
- *
- * Returns:    EBADF                   The file descriptor is invalid
- *             EBADF                   The file descriptor is not a socket
- *             0                       Success
- *
- * Implicit returns:
- *             *resultfp (modified)            Fileproc pointer
- *             *resultpipe (modified)          pipe pointer
- *
- * Notes:      The second EBADF should probably be something else to make
- *             the error condition distinct.
- */
-int
-fp_getfpipe(proc_t p, int fd, struct fileproc **resultfp,
-           struct pipe **resultpipe)
-{
-       struct filedesc *fdp = p->p_fd;
-       struct fileproc *fp;
-
-       proc_fdlock_spin(p);
-       if (fd < 0 || fd >= fdp->fd_nfiles ||
-                       (fp = fdp->fd_ofiles[fd]) == NULL ||
-                       (fdp->fd_ofileflags[fd] & UF_RESERVED)) {
-               proc_fdunlock(p);
-               return (EBADF);
-       }
-       if (fp->f_type != DTYPE_PIPE) {
-               proc_fdunlock(p);
-               return(EBADF);
-       }
-       fp->f_iocount++;
-
-       if (resultfp)
-               *resultfp = fp;
-       if (resultpipe)
-               *resultpipe = (struct pipe *)fp->f_data;
-       proc_fdunlock(p);
-
-       return (0);
-}
-
-
-/*
- * fp_lookup
- *
- * Description:        Get fileproc pointer for a given fd from the per process
- *             open file table of the specified process and if successful,
- *             increment the f_iocount
- *
- * Parameters: p                               Process in which fd lives
- *             fd                              fd to get information for
- *             resultfp                        Pointer to result fileproc
- *                                             pointer area, or 0 if none
- *             locked                          !0 if the caller holds the
- *                                             proc_fdlock, 0 otherwise
- *
- * Returns:    0                       Success
- *             EBADF                   Bad file descriptor
- *
- * Implicit returns:
- *             *resultfp (modified)            Fileproc pointer
- *
- * Locks:      If the argument 'locked' is non-zero, then the caller is
- *             expected to have taken and held the proc_fdlock; if it is
- *             zero, than this routine internally takes and drops this lock.
- */
-int
-fp_lookup(proc_t p, int fd, struct fileproc **resultfp, int locked)
-{
-       struct filedesc *fdp = p->p_fd;
-       struct fileproc *fp;
-
-       if (!locked)
-               proc_fdlock_spin(p);
-       if (fd < 0 || fdp == NULL || fd >= fdp->fd_nfiles ||
-                       (fp = fdp->fd_ofiles[fd]) == NULL ||
-                       (fdp->fd_ofileflags[fd] & UF_RESERVED)) {
-               if (!locked)
-                       proc_fdunlock(p);
-               return (EBADF);
-       }
-       fp->f_iocount++;
-
-       if (resultfp)
-               *resultfp = fp;
-       if (!locked)
-               proc_fdunlock(p);
-
-       return (0);
-}
-
-
-/*
- * fp_tryswap
+ * fp_tryswap
   *
   * Description: Swap the fileproc pointer for a given fd with a new
   *             fileproc pointer in the per-process open file table of
   *             the specified process.  The fdlock must be held at entry.
   *
   * Description: Swap the fileproc pointer for a given fd with a new
   *             fileproc pointer in the per-process open file table of
   *             the specified process.  The fdlock must be held at entry.
+ *             Iff the swap is successful, the old fileproc pointer is freed.
   *
   * Parameters:  p              Process containing the fd
   *             fd              The fd of interest
   *
   * Parameters:  p              Process containing the fd
   *             fd              The fd of interest
@@ -3943,7 +4138,7 @@ fp_lookup(proc_t p, int fd, struct fileproc **resultfp, int locked)
   * Returns:    0               Success
   *             EBADF           Bad file descriptor
   *             EINTR           Interrupted
   * Returns:    0               Success
   *             EBADF           Bad file descriptor
   *             EINTR           Interrupted
- *             EKEEPLOOKING    f_iocount changed while lock was dropped.
+ *             EKEEPLOOKING    Other references were active, try again.
   */
  int
  fp_tryswap(proc_t p, int fd, struct fileproc *nfp)
   */
  int
  fp_tryswap(proc_t p, int fd, struct fileproc *nfp)
@@ -3953,27 +4148,35 @@ fp_tryswap(proc_t p, int fd, struct fileproc *nfp)
  
         proc_fdlock_assert(p, LCK_MTX_ASSERT_OWNED);
  
  
         proc_fdlock_assert(p, LCK_MTX_ASSERT_OWNED);
  
-       if (0 != (error = fp_lookup(p, fd, &fp, 1)))
-               return (error);
+       if (0 != (error = fp_lookup(p, fd, &fp, 1))) {
+               return error;
+       }
         /*
          * At this point, our caller (change_guardedfd_np) has
         /*
          * At this point, our caller (change_guardedfd_np) has
-        * one f_iocount reference, and we just took another
+        * one fp_iocount reference, and we just took another
          * one to begin the replacement.
          * one to begin the replacement.
+        * fp and nfp have a +1 reference from allocation.
+        * Thus if no-one else is looking, fp_iocount should be 3.
          */
          */
-       if (fp->f_iocount < 2) {
-               panic("f_iocount too small %d", fp->f_iocount);
-       } else if (2 == fp->f_iocount) {
-
+       if (os_ref_get_count(&fp->fp_iocount) < 3 ||
+           1 != os_ref_get_count(&nfp->fp_iocount)) {
+               panic("%s: fp_iocount", __func__);
+       } else if (3 == os_ref_get_count(&fp->fp_iocount)) {
                 /* Copy the contents of *fp, preserving the "type" of *nfp */
  
                 /* Copy the contents of *fp, preserving the "type" of *nfp */
  
-               nfp->f_flags = (nfp->f_flags & FP_TYPEMASK) |
-                       (fp->f_flags & ~FP_TYPEMASK);
-               nfp->f_iocount = fp->f_iocount;
-               nfp->f_fglob = fp->f_fglob;
-               nfp->f_wset = fp->f_wset;
+               nfp->fp_flags = (nfp->fp_flags & FP_TYPEMASK) |
+                   (fp->fp_flags & ~FP_TYPEMASK);
+               os_ref_retain_locked(&nfp->fp_iocount);
+               os_ref_retain_locked(&nfp->fp_iocount);
+               nfp->fp_glob = fp->fp_glob;
+               nfp->fp_wset = fp->fp_wset;
  
                 p->p_fd->fd_ofiles[fd] = nfp;
  
                 p->p_fd->fd_ofiles[fd] = nfp;
-               (void) fp_drop(p, fd, nfp, 1);
+               fp_drop(p, fd, nfp, 1);
+
+               os_ref_release_live(&fp->fp_iocount);
+               os_ref_release_live(&fp->fp_iocount);
+               fileproc_free(fp);
         } else {
                 /*
                  * Wait for all other active references to evaporate.
         } else {
                 /*
                  * Wait for all other active references to evaporate.
@@ -3987,83 +4190,10 @@ fp_tryswap(proc_t p, int fd, struct fileproc *nfp)
                          * reevaluation of the change-guard attempt.
                          */
                         error = EKEEPLOOKING;
                          * reevaluation of the change-guard attempt.
                          */
                         error = EKEEPLOOKING;
-                       printf("%s: lookup collision fd %d\n", __func__, fd);
                 }
                 (void) fp_drop(p, fd, fp, 1);
         }
                 }
                 (void) fp_drop(p, fd, fp, 1);
         }
-       return (error);
-}
-
-
-/*
- * fp_drop_written
- *
- * Description:        Set the FP_WRITTEN flag on the fileproc and drop the I/O
- *             reference previously taken by calling fp_lookup et. al.
- *
- * Parameters: p                               Process in which the fd lives
- *             fd                              fd associated with the fileproc
- *             fp                              fileproc on which to set the
- *                                             flag and drop the reference
- *
- * Returns:    0                               Success
- *     fp_drop:EBADF                           Bad file descriptor
- *
- * Locks:      This function internally takes and drops the proc_fdlock for
- *             the supplied process
- *
- * Notes:      The fileproc must correspond to the fd in the supplied proc
- */
-int
-fp_drop_written(proc_t p, int fd, struct fileproc *fp)
-{
-        int error;
-
-       proc_fdlock_spin(p);
-
-       fp->f_flags |= FP_WRITTEN;
-
-       error = fp_drop(p, fd, fp, 1);
-
-       proc_fdunlock(p);
-
-       return (error);
-}
-
-
-/*
- * fp_drop_event
- *
- * Description:        Set the FP_WAITEVENT flag on the fileproc and drop the I/O
- *             reference previously taken by calling fp_lookup et. al.
- *
- * Parameters: p                               Process in which the fd lives
- *             fd                              fd associated with the fileproc
- *             fp                              fileproc on which to set the
- *                                             flag and drop the reference
- *
- * Returns:    0                               Success
- *     fp_drop:EBADF                           Bad file descriptor
- *
- * Locks:      This function internally takes and drops the proc_fdlock for
- *             the supplied process
- *
- * Notes:      The fileproc must correspond to the fd in the supplied proc
- */
-int
-fp_drop_event(proc_t p, int fd, struct fileproc *fp)
-{
-        int error;
-
-       proc_fdlock_spin(p);
-
-       fp->f_flags |= FP_WAITEVENT;
-
-       error = fp_drop(p, fd, fp, 1);
-
-       proc_fdunlock(p);
-
-       return (error);
+       return error;
  }
  
  
  }
  
  
@@ -4094,35 +4224,39 @@ int
  fp_drop(proc_t p, int fd, struct fileproc *fp, int locked)
  {
         struct filedesc *fdp = p->p_fd;
  fp_drop(proc_t p, int fd, struct fileproc *fp, int locked)
  {
         struct filedesc *fdp = p->p_fd;
-       int     needwakeup = 0;
+       int     needwakeup = 0;
  
  
-       if (!locked)
+       if (!locked) {
                 proc_fdlock_spin(p);
                 proc_fdlock_spin(p);
-        if ((fp == FILEPROC_NULL) && (fd < 0 || fd >= fdp->fd_nfiles ||
-                       (fp = fdp->fd_ofiles[fd]) == NULL ||
-                       ((fdp->fd_ofileflags[fd] & UF_RESERVED) &&
-                        !(fdp->fd_ofileflags[fd] & UF_CLOSING)))) {
-               if (!locked)
+       }
+       if ((fp == FILEPROC_NULL) && (fd < 0 || fd >= fdp->fd_nfiles ||
+           (fp = fdp->fd_ofiles[fd]) == NULL ||
+           ((fdp->fd_ofileflags[fd] & UF_RESERVED) &&
+           !(fdp->fd_ofileflags[fd] & UF_CLOSING)))) {
+               if (!locked) {
                         proc_fdunlock(p);
                         proc_fdunlock(p);
-               return (EBADF);
+               }
+               return EBADF;
         }
         }
-       fp->f_iocount--;
  
  
-       if (fp->f_iocount == 0) {
-               if (fp->f_flags & FP_SELCONFLICT)
-                       fp->f_flags &= ~FP_SELCONFLICT;
+       if (1 == os_ref_release_locked(&fp->fp_iocount)) {
+               if (fp->fp_flags & FP_SELCONFLICT) {
+                       fp->fp_flags &= ~FP_SELCONFLICT;
+               }
  
                 if (p->p_fpdrainwait) {
                         p->p_fpdrainwait = 0;
                         needwakeup = 1;
                 }
         }
  
                 if (p->p_fpdrainwait) {
                         p->p_fpdrainwait = 0;
                         needwakeup = 1;
                 }
         }
-       if (!locked)
+       if (!locked) {
                 proc_fdunlock(p);
                 proc_fdunlock(p);
-       if (needwakeup)
-               wakeup(&p->p_fpdrainwait);
+       }
+       if (needwakeup) {
+               wakeup(&p->p_fpdrainwait);
+       }
  
  
-       return (0);
+       return 0;
  }
  
  
  }
  
  
@@ -4146,7 +4280,7 @@ fp_drop(proc_t p, int fd, struct fileproc *fp, int locked)
   * Locks:      This function internally takes and drops the proc_fdlock for
   *             the current process
   *
   * Locks:      This function internally takes and drops the proc_fdlock for
   *             the current process
   *
- * Notes:      If successful, this function increments the f_iocount on the
+ * Notes:      If successful, this function increments the fp_iocount on the
   *             fd's corresponding fileproc.
   *
   *             The fileproc referenced is not returned; because of this, care
   *             fd's corresponding fileproc.
   *
   *             The fileproc referenced is not returned; because of this, care
@@ -4155,7 +4289,7 @@ fp_drop(proc_t p, int fd, struct fileproc *fp, int locked)
   *             not be recoverable from the vnode, if there is a subsequent
   *             close that destroys the associate fileproc.  The caller should
   *             therefore retain their own reference on the fileproc so that
   *             not be recoverable from the vnode, if there is a subsequent
   *             close that destroys the associate fileproc.  The caller should
   *             therefore retain their own reference on the fileproc so that
- *             the f_iocount can be dropped subsequently.  Failure to do this
+ *             the fp_iocount can be dropped subsequently.  Failure to do this
   *             can result in the returned pointer immediately becoming invalid
   *             following the call.
   *
   *             can result in the returned pointer immediately becoming invalid
   *             following the call.
   *
@@ -4164,28 +4298,9 @@ fp_drop(proc_t p, int fd, struct fileproc *fp, int locked)
  int
  file_vnode(int fd, struct vnode **vpp)
  {
  int
  file_vnode(int fd, struct vnode **vpp)
  {
-       proc_t p = current_proc();
-       struct fileproc *fp;
-       int error;
-
-       proc_fdlock_spin(p);
-       if ( (error = fp_lookup(p, fd, &fp, 1)) ) {
-               proc_fdunlock(p);
-               return(error);
-       }
-       if (fp->f_type != DTYPE_VNODE) {
-               fp_drop(p, fd, fp,1);
-               proc_fdunlock(p);
-               return(EINVAL);
-       }
-       if (vpp != NULL)
-               *vpp = (struct vnode *)fp->f_data;
-       proc_fdunlock(p);
-
-       return(0);
+       return file_vnode_withvid(fd, vpp, NULL);
  }
  
  }
  
-
  /*
   * file_vnode_withvid
   *
  /*
   * file_vnode_withvid
   *
@@ -4207,7 +4322,7 @@ file_vnode(int fd, struct vnode **vpp)
   * Locks:      This function internally takes and drops the proc_fdlock for
   *             the current process
   *
   * Locks:      This function internally takes and drops the proc_fdlock for
   *             the current process
   *
- * Notes:      If successful, this function increments the f_iocount on the
+ * Notes:      If successful, this function increments the fp_iocount on the
   *             fd's corresponding fileproc.
   *
   *             The fileproc referenced is not returned; because of this, care
   *             fd's corresponding fileproc.
   *
   *             The fileproc referenced is not returned; because of this, care
@@ -4216,43 +4331,30 @@ file_vnode(int fd, struct vnode **vpp)
   *             not be recoverable from the vnode, if there is a subsequent
   *             close that destroys the associate fileproc.  The caller should
   *             therefore retain their own reference on the fileproc so that
   *             not be recoverable from the vnode, if there is a subsequent
   *             close that destroys the associate fileproc.  The caller should
   *             therefore retain their own reference on the fileproc so that
- *             the f_iocount can be dropped subsequently.  Failure to do this
+ *             the fp_iocount can be dropped subsequently.  Failure to do this
   *             can result in the returned pointer immediately becoming invalid
   *             following the call.
   *
   *             Use of this function is discouraged.
   */
  int
   *             can result in the returned pointer immediately becoming invalid
   *             following the call.
   *
   *             Use of this function is discouraged.
   */
  int
-file_vnode_withvid(int fd, struct vnode **vpp, uint32_t * vidp)
+file_vnode_withvid(int fd, struct vnode **vpp, uint32_t *vidp)
  {
  {
-       proc_t p = current_proc();
         struct fileproc *fp;
         struct fileproc *fp;
-       vnode_t vp;
         int error;
  
         int error;
  
-       proc_fdlock_spin(p);
-       if ( (error = fp_lookup(p, fd, &fp, 1)) ) {
-               proc_fdunlock(p);
-               return(error);
-       }
-       if (fp->f_type != DTYPE_VNODE) {
-               fp_drop(p, fd, fp,1);
-               proc_fdunlock(p);
-               return(EINVAL);
+       error = fp_get_ftype(current_proc(), fd, DTYPE_VNODE, EINVAL, &fp);
+       if (error == 0) {
+               if (vpp) {
+                       *vpp = fp->f_data;
+               }
+               if (vidp) {
+                       *vidp = vnode_vid(fp->f_data);
+               }
         }
         }
-       vp = (struct vnode *)fp->f_data;
-       if (vpp != NULL)
-               *vpp = vp;
-
-       if ((vidp != NULL) && (vp != NULLVP))
-               *vidp = (uint32_t)vp->v_id;
-
-       proc_fdunlock(p);
-
-       return(0);
+       return error;
  }
  
  }
  
-
  /*
   * file_socket
   *
  /*
   * file_socket
   *
@@ -4272,7 +4374,7 @@ file_vnode_withvid(int fd, struct vnode **vpp, uint32_t * vidp)
   * Locks:      This function internally takes and drops the proc_fdlock for
   *             the current process
   *
   * Locks:      This function internally takes and drops the proc_fdlock for
   *             the current process
   *
- * Notes:      If successful, this function increments the f_iocount on the
+ * Notes:      If successful, this function increments the fp_iocount on the
   *             fd's corresponding fileproc.
   *
   *             The fileproc referenced is not returned; because of this, care
   *             fd's corresponding fileproc.
   *
   *             The fileproc referenced is not returned; because of this, care
@@ -4281,7 +4383,7 @@ file_vnode_withvid(int fd, struct vnode **vpp, uint32_t * vidp)
   *             not be recoverable from the socket, if there is a subsequent
   *             close that destroys the associate fileproc.  The caller should
   *             therefore retain their own reference on the fileproc so that
   *             not be recoverable from the socket, if there is a subsequent
   *             close that destroys the associate fileproc.  The caller should
   *             therefore retain their own reference on the fileproc so that
- *             the f_iocount can be dropped subsequently.  Failure to do this
+ *             the fp_iocount can be dropped subsequently.  Failure to do this
   *             can result in the returned pointer immediately becoming invalid
   *             following the call.
   *
   *             can result in the returned pointer immediately becoming invalid
   *             following the call.
   *
@@ -4290,24 +4392,16 @@ file_vnode_withvid(int fd, struct vnode **vpp, uint32_t * vidp)
  int
  file_socket(int fd, struct socket **sp)
  {
  int
  file_socket(int fd, struct socket **sp)
  {
-       proc_t p = current_proc();
         struct fileproc *fp;
         int error;
  
         struct fileproc *fp;
         int error;
  
-       proc_fdlock_spin(p);
-       if ( (error = fp_lookup(p, fd, &fp, 1)) ) {
-               proc_fdunlock(p);
-               return(error);
-       }
-       if (fp->f_type != DTYPE_SOCKET) {
-               fp_drop(p, fd, fp,1);
-               proc_fdunlock(p);
-               return(ENOTSOCK);
+       error = fp_get_ftype(current_proc(), fd, DTYPE_SOCKET, ENOTSOCK, &fp);
+       if (error == 0) {
+               if (sp) {
+                       *sp = (struct socket *)fp->f_data;
+               }
         }
         }
-       *sp = (struct socket *)fp->f_data;
-       proc_fdunlock(p);
-
-       return(0);
+       return error;
  }
  
  
  }
  
  
@@ -4330,28 +4424,23 @@ file_socket(int fd, struct socket **sp)
   *
   * Locks:      This function internally takes and drops the proc_fdlock for
   *             the current process
   *
   * Locks:      This function internally takes and drops the proc_fdlock for
   *             the current process
- *
- * Notes:      This function will internally increment and decrement the
- *             f_iocount of the fileproc as part of its operation.
   */
  int
  file_flags(int fd, int *flags)
  {
   */
  int
  file_flags(int fd, int *flags)
  {
-
         proc_t p = current_proc();
         struct fileproc *fp;
         proc_t p = current_proc();
         struct fileproc *fp;
-       int error;
+       int error = EBADF;
  
         proc_fdlock_spin(p);
  
         proc_fdlock_spin(p);
-       if ( (error = fp_lookup(p, fd, &fp, 1)) ) {
-               proc_fdunlock(p);
-               return(error);
+       fp = fp_get_noref_locked(p, fd);
+       if (fp) {
+               *flags = (int)fp->f_flag;
+               error = 0;
         }
         }
-       *flags = (int)fp->f_flag;
-       fp_drop(p, fd, fp,1);
         proc_fdunlock(p);
  
         proc_fdunlock(p);
  
-       return(0);
+       return error;
  }
  
  
  }
  
  
@@ -4366,28 +4455,17 @@ file_flags(int fd, int *flags)
   *                                             to be dropped
   *
   * Returns:    0                               Success
   *                                             to be dropped
   *
   * Returns:    0                               Success
- *             EBADF                           Bad file descriptor
   *
   * Description:        Given an fd, look it up in the current process's per process
   *
   * Description:        Given an fd, look it up in the current process's per process
- *             open file table, and drop it's fileproc's f_iocount by one
+ *             open file table, and drop it's fileproc's fp_iocount by one
   *
   * Notes:      This is intended as a corresponding operation to the functions
   *             file_vnode() and file_socket() operations.
   *
   *
   * Notes:      This is intended as a corresponding operation to the functions
   *             file_vnode() and file_socket() operations.
   *
- *             Technically, the close reference is supposed to be protected
- *             by a fileproc_drain(), however, a drain will only block if
- *             the fd refers to a character device, and that device has had
- *             preparefileread() called on it.  If it refers to something
- *             other than a character device, then the drain will occur and
- *             block each close attempt, rather than merely the last close.
- *
- *             Since it's possible for an fd that refers to a character
- *             device to have an intermediate close followed by an open to
- *             cause a different file to correspond to that descriptor,
- *             unless there was a cautionary reference taken on the fileproc,
- *             this is an inherently unsafe function.  This happens in the
- *             case where multiple fd's in a process refer to the same
- *             character device (e.g. stdin/out/err pointing to a tty, etc.).
+ *             If the caller can't possibly hold an I/O reference,
+ *             this function will panic the kernel rather than allowing
+ *             for memory corruption. Callers should always call this
+ *             because they acquired an I/O reference on this file before.
   *
   *             Use of this function is discouraged.
   */
   *
   *             Use of this function is discouraged.
   */
@@ -4396,21 +4474,15 @@ file_drop(int fd)
  {
         struct fileproc *fp;
         proc_t p = current_proc();
  {
         struct fileproc *fp;
         proc_t p = current_proc();
-       int     needwakeup = 0;
+       int     needwakeup = 0;
  
         proc_fdlock_spin(p);
  
         proc_fdlock_spin(p);
-       if (fd < 0 || fd >= p->p_fd->fd_nfiles ||
-                       (fp = p->p_fd->fd_ofiles[fd]) == NULL ||
-                       ((p->p_fd->fd_ofileflags[fd] & UF_RESERVED) &&
-                        !(p->p_fd->fd_ofileflags[fd] & UF_CLOSING))) {
-               proc_fdunlock(p);
-               return (EBADF);
-       }
-       fp->f_iocount --;
+       fp = fp_get_noref_locked_with_iocount(p, fd);
  
  
-       if (fp->f_iocount == 0) {
-               if (fp->f_flags & FP_SELCONFLICT)
-                       fp->f_flags &= ~FP_SELCONFLICT;
+       if (1 == os_ref_release_locked(&fp->fp_iocount)) {
+               if (fp->fp_flags & FP_SELCONFLICT) {
+                       fp->fp_flags &= ~FP_SELCONFLICT;
+               }
  
                 if (p->p_fpdrainwait) {
                         p->p_fpdrainwait = 0;
  
                 if (p->p_fpdrainwait) {
                         p->p_fpdrainwait = 0;
@@ -4419,77 +4491,16 @@ file_drop(int fd)
         }
         proc_fdunlock(p);
  
         }
         proc_fdunlock(p);
  
-       if (needwakeup)
-               wakeup(&p->p_fpdrainwait);
-       return(0);
+       if (needwakeup) {
+               wakeup(&p->p_fpdrainwait);
+       }
+       return 0;
  }
  
  
  }
  
  
-static int falloc_withalloc_locked(proc_t, struct fileproc **, int *,
-    vfs_context_t, struct fileproc * (*)(void *), void *, int);
  
  /*
  
  /*
- * falloc
- *
- * Description:        Allocate an entry in the per process open file table and
- *             return the corresponding fileproc and fd.
- *
- * Parameters: p                               The process in whose open file
- *                                             table the fd is to be allocated
- *             resultfp                        Pointer to fileproc pointer
- *                                             return area
- *             resultfd                        Pointer to fd return area
- *             ctx                             VFS context
- *
- * Returns:    0                               Success
- *     falloc:ENFILE                           Too many open files in system
- *     falloc:EMFILE                           Too many open files in process
- *     falloc:ENOMEM                           M_FILEPROC or M_FILEGLOB zone
- *                                             exhausted
- *
- * Implicit returns:
- *             *resultfd (modified)            Returned fileproc pointer
- *             *resultfd (modified)            Returned fd
- *
- * Locks:      This function takes and drops the proc_fdlock; if this lock
- *             is already held, use falloc_locked() instead.
- *
- * Notes:      This function takes separate process and context arguments
- *             solely to support kern_exec.c; otherwise, it would take
- *             neither, and expect falloc_locked() to use the
- *             vfs_context_current() routine internally.
- */
-int
-falloc(proc_t p, struct fileproc **resultfp, int *resultfd, vfs_context_t ctx)
-{
-       return (falloc_withalloc(p, resultfp, resultfd, ctx,
-           fileproc_alloc_init, NULL));
-}
-
-/*
- * Like falloc, but including the fileproc allocator and create-args
- */
-int
-falloc_withalloc(proc_t p, struct fileproc **resultfp, int *resultfd,
-    vfs_context_t ctx, fp_allocfn_t fp_zalloc, void *arg)
-{
-       int error;
-
-       proc_fdlock(p);
-       error = falloc_withalloc_locked(p,
-           resultfp, resultfd, ctx, fp_zalloc, arg, 1);
-       proc_fdunlock(p);
-
-       return (error);
-}
-
-/*
- * "uninitialized" ops -- ensure fg->fg_ops->fo_type always exists
- */
-static const struct fileops uninitops;
-
-/*
- * falloc_locked
+ * falloc_withalloc
   *
   * Create a new open file structure and allocate
   * a file descriptor for the process that refers to it.
   *
   * Create a new open file structure and allocate
   * a file descriptor for the process that refers to it.
@@ -4505,67 +4516,51 @@ static const struct fileops uninitops;
   *                                             return area
   *             resultfd                        Pointer to fd return area
   *             ctx                             VFS context
   *                                             return area
   *             resultfd                        Pointer to fd return area
   *             ctx                             VFS context
- *             locked                          Flag to indicate whether the
- *                                             caller holds proc_fdlock
+ *             fp_zalloc                       fileproc allocator to use
+ *             crarg                           allocator args
   *
   * Returns:    0                               Success
   *             ENFILE                          Too many open files in system
   *             fdalloc:EMFILE                  Too many open files in process
   *
   * Returns:    0                               Success
   *             ENFILE                          Too many open files in system
   *             fdalloc:EMFILE                  Too many open files in process
- *             ENOMEM                          M_FILEPROC or M_FILEGLOB zone
+ *             fdalloc:ENOMEM                  M_OFILETABL zone exhausted
+ *             ENOMEM                          fp_zone or fg_zone zone
   *                                             exhausted
   *                                             exhausted
- *     fdalloc:ENOMEM
   *
   * Implicit returns:
   *             *resultfd (modified)            Returned fileproc pointer
   *             *resultfd (modified)            Returned fd
   *
   *
   * Implicit returns:
   *             *resultfd (modified)            Returned fileproc pointer
   *             *resultfd (modified)            Returned fd
   *
- * Locks:      If the parameter 'locked' is zero, this function takes and
- *             drops the proc_fdlock; if non-zero, the caller must hold the
- *             lock.
- *
- * Notes:      If you intend to use a non-zero 'locked' parameter, use the
- *             utility function falloc() instead.
- *
- *             This function takes separate process and context arguments
+ * Notes:      This function takes separate process and context arguments
   *             solely to support kern_exec.c; otherwise, it would take
   *             neither, and use the vfs_context_current() routine internally.
   */
  int
   *             solely to support kern_exec.c; otherwise, it would take
   *             neither, and use the vfs_context_current() routine internally.
   */
  int
-falloc_locked(proc_t p, struct fileproc **resultfp, int *resultfd,
-             vfs_context_t ctx, int locked)
-{
-       return (falloc_withalloc_locked(p, resultfp, resultfd, ctx,
-           fileproc_alloc_init, NULL, locked));
-}
-
-static int
-falloc_withalloc_locked(proc_t p, struct fileproc **resultfp, int *resultfd,
-    vfs_context_t ctx, fp_allocfn_t fp_zalloc, void *crarg,
-    int locked)
+falloc_withalloc(proc_t p, struct fileproc **resultfp, int *resultfd,
+    vfs_context_t ctx, fp_allocfn_t fp_zalloc, void *crarg)
  {
         struct fileproc *fp;
         struct fileglob *fg;
         int error, nfd;
  
  {
         struct fileproc *fp;
         struct fileglob *fg;
         int error, nfd;
  
-       if (!locked)
-               proc_fdlock(p);
-       if ( (error = fdalloc(p, 0, &nfd)) ) {
-               if (!locked)
-                       proc_fdunlock(p);
-               return (error);
-       }
+       /* Make sure we don't go beyond the system-wide limit */
         if (nfiles >= maxfiles) {
         if (nfiles >= maxfiles) {
-               if (!locked)
-                       proc_fdunlock(p);
                 tablefull("file");
                 tablefull("file");
-               return (ENFILE);
+               return ENFILE;
         }
         }
+
+       proc_fdlock(p);
+
+       /* fdalloc will make sure the process stays below per-process limit */
+       if ((error = fdalloc(p, 0, &nfd))) {
+               proc_fdunlock(p);
+               return error;
+       }
+
  #if CONFIG_MACF
         error = mac_file_check_create(proc_ucred(p));
         if (error) {
  #if CONFIG_MACF
         error = mac_file_check_create(proc_ucred(p));
         if (error) {
-               if (!locked)
-                       proc_fdunlock(p);
-               return (error);
+               proc_fdunlock(p);
+               return error;
         }
  #endif
  
         }
  #endif
  
@@ -4579,89 +4574,52 @@ falloc_withalloc_locked(proc_t p, struct fileproc **resultfp, int *resultfd,
  
         fp = (*fp_zalloc)(crarg);
         if (fp == NULL) {
  
         fp = (*fp_zalloc)(crarg);
         if (fp == NULL) {
-               if (locked)
-                       proc_fdlock(p);
-               return (ENOMEM);
+               return ENOMEM;
         }
         }
-       MALLOC_ZONE(fg, struct fileglob *, sizeof(struct fileglob), M_FILEGLOB, M_WAITOK);
-       if (fg == NULL) {
-               fileproc_free(fp);
-               if (locked)
-                       proc_fdlock(p);
-               return (ENOMEM);
-       }
-       bzero(fg, sizeof(struct fileglob));
-       lck_mtx_init(&fg->fg_lock, file_lck_grp, file_lck_attr);
+       fg = zalloc_flags(fg_zone, Z_WAITOK | Z_ZERO);
+       lck_mtx_init(&fg->fg_lock, &file_lck_grp, LCK_ATTR_NULL);
  
  
-       fp->f_iocount = 1;
-       fg->fg_count = 1;
+       os_ref_retain_locked(&fp->fp_iocount);
+       os_ref_init_raw(&fg->fg_count, &f_refgrp);
         fg->fg_ops = &uninitops;
         fg->fg_ops = &uninitops;
-       fp->f_fglob = fg;
+       fp->fp_glob = fg;
  #if CONFIG_MACF
         mac_file_label_init(fg);
  #endif
  
         kauth_cred_ref(ctx->vc_ucred);
  
  #if CONFIG_MACF
         mac_file_label_init(fg);
  #endif
  
         kauth_cred_ref(ctx->vc_ucred);
  
-       proc_fdlock(p);
-
         fp->f_cred = ctx->vc_ucred;
  
  #if CONFIG_MACF
         mac_file_label_associate(fp->f_cred, fg);
  #endif
  
         fp->f_cred = ctx->vc_ucred;
  
  #if CONFIG_MACF
         mac_file_label_associate(fp->f_cred, fg);
  #endif
  
-       OSAddAtomic(1, &nfiles);
+       os_atomic_inc(&nfiles, relaxed);
+
+       proc_fdlock(p);
  
         p->p_fd->fd_ofiles[nfd] = fp;
  
  
         p->p_fd->fd_ofiles[nfd] = fp;
  
-       if (!locked)
-               proc_fdunlock(p);
+       proc_fdunlock(p);
  
  
-       if (resultfp)
+       if (resultfp) {
                 *resultfp = fp;
                 *resultfp = fp;
-       if (resultfd)
+       }
+       if (resultfd) {
                 *resultfd = nfd;
                 *resultfd = nfd;
+       }
  
  
-       return (0);
+       return 0;
  }
  
  }
  
-
-/*
- * fg_free
- *
- * Description:        Free a file structure; drop the global open file count, and
- *             drop the credential reference, if the fileglob has one, and
- *             destroy the instance mutex before freeing
- *
- * Parameters: fg                              Pointer to fileglob to be
- *                                             freed
- *
- * Returns:    void
- */
-void
-fg_free(struct fileglob *fg)
+int
+falloc(proc_t p, struct fileproc **resultfp, int *resultfd, vfs_context_t ctx)
  {
  {
-       OSAddAtomic(-1, &nfiles);
-
-       if (fg->fg_vn_data) {
-               fg_vn_data_free(fg->fg_vn_data);
-               fg->fg_vn_data = NULL;
-       }
-
-       if (IS_VALID_CRED(fg->fg_cred)) {
-               kauth_cred_unref(&fg->fg_cred);
-       }
-       lck_mtx_destroy(&fg->fg_lock, file_lck_grp);
-
-#if CONFIG_MACF
-       mac_file_label_destroy(fg);
-#endif
-       FREE_ZONE(fg, sizeof *fg, M_FILEGLOB);
+       return falloc_withalloc(p, resultfp, resultfd, ctx,
+                  fileproc_alloc_init, NULL);
  }
  
  }
  
-
-
  /*
   * fdexec
   *
  /*
   * fdexec
   *
@@ -4690,7 +4648,7 @@ fdexec(proc_t p, short flags, int self_exec)
         boolean_t cloexec_default = (flags & POSIX_SPAWN_CLOEXEC_DEFAULT) != 0;
         thread_t self = current_thread();
         struct uthread *ut = get_bsdthread_info(self);
         boolean_t cloexec_default = (flags & POSIX_SPAWN_CLOEXEC_DEFAULT) != 0;
         thread_t self = current_thread();
         struct uthread *ut = get_bsdthread_info(self);
-       struct kqueue *dealloc_kq = NULL;
+       struct kqworkq *dealloc_kqwq = NULL;
  
         /*
          * If the current thread is bound as a workq/workloop
  
         /*
          * If the current thread is bound as a workq/workloop
@@ -4708,11 +4666,10 @@ fdexec(proc_t p, short flags, int self_exec)
          * subsequent kqueue closes go faster.
          */
         knotes_dealloc(p);
          * subsequent kqueue closes go faster.
          */
         knotes_dealloc(p);
-       assert(fdp->fd_knlistsize == -1);
+       assert(fdp->fd_knlistsize == 0);
         assert(fdp->fd_knhashmask == 0);
  
         for (i = fdp->fd_lastfile; i >= 0; i--) {
         assert(fdp->fd_knhashmask == 0);
  
         for (i = fdp->fd_lastfile; i >= 0; i--) {
-
                 struct fileproc *fp = fdp->fd_ofiles[i];
                 char *flagp = &fdp->fd_ofileflags[i];
  
                 struct fileproc *fp = fdp->fd_ofiles[i];
                 char *flagp = &fdp->fd_ofileflags[i];
  
@@ -4723,52 +4680,35 @@ fdexec(proc_t p, short flags, int self_exec)
                          * except files marked explicitly as "inherit" and
                          * not marked close-on-exec.
                          */
                          * except files marked explicitly as "inherit" and
                          * not marked close-on-exec.
                          */
-                       if ((*flagp & (UF_EXCLOSE|UF_INHERIT)) != UF_INHERIT)
+                       if ((*flagp & (UF_EXCLOSE | UF_INHERIT)) != UF_INHERIT) {
                                 *flagp |= UF_EXCLOSE;
                                 *flagp |= UF_EXCLOSE;
+                       }
                         *flagp &= ~UF_INHERIT;
                 }
  
                 if (
                         *flagp &= ~UF_INHERIT;
                 }
  
                 if (
-                   ((*flagp & (UF_RESERVED|UF_EXCLOSE)) == UF_EXCLOSE)
+                       ((*flagp & (UF_RESERVED | UF_EXCLOSE)) == UF_EXCLOSE)
  #if CONFIG_MACF
  #if CONFIG_MACF
-                   || (fp && mac_file_check_inherit(proc_ucred(p), fp->f_fglob))
+                       || (fp && mac_file_check_inherit(proc_ucred(p), fp->fp_glob))
  #endif
  #endif
-               ) {
-                       procfdtbl_clearfd(p, i);
-                       if (i == fdp->fd_lastfile && i > 0)
-                               fdp->fd_lastfile--;
-                       if (i < fdp->fd_freefile)
-                               fdp->fd_freefile = i;
-
-                       /*
-                        * Wait for any third party viewers (e.g., lsof)
-                        * to release their references to this fileproc.
-                        */
-                       while (fp->f_iocount > 0) {
-                               p->p_fpdrainwait = 1;
-                               msleep(&p->p_fpdrainwait, &p->p_fdmlock, PRIBIO,
-                                   "fpdrain", NULL);
-                       }
-                       if (fp->f_flags & FP_WAITEVENT) {
-                               (void)waitevent_close(p, fp);
-                       }
-                       closef_locked(fp, fp->f_fglob, p);
-
-                       fileproc_free(fp);
+                       ) {
+                       fp_close_and_unlock(p, i, fp, 0);
+                       proc_fdlock(p);
                 }
         }
  
         /* release the per-process workq kq */
         if (fdp->fd_wqkqueue) {
                 }
         }
  
         /* release the per-process workq kq */
         if (fdp->fd_wqkqueue) {
-               dealloc_kq = fdp->fd_wqkqueue;
+               dealloc_kqwq = fdp->fd_wqkqueue;
                 fdp->fd_wqkqueue = NULL;
         }
                 fdp->fd_wqkqueue = NULL;
         }
-          
+
         proc_fdunlock(p);
  
         /* Anything to free? */
         proc_fdunlock(p);
  
         /* Anything to free? */
-       if (dealloc_kq)
-               kqueue_dealloc(dealloc_kq);
+       if (dealloc_kqwq) {
+               kqworkq_dealloc(dealloc_kqwq);
+       }
  }
  
  
  }
  
  
@@ -4814,12 +4754,9 @@ fdcopy(proc_t p, vnode_t uth_cdir)
         struct filedesc *newfdp, *fdp = p->p_fd;
         int i;
         struct fileproc *ofp, *fp;
         struct filedesc *newfdp, *fdp = p->p_fd;
         int i;
         struct fileproc *ofp, *fp;
-       vnode_t v_dir;
+       vnode_t v_dir;
  
  
-       MALLOC_ZONE(newfdp, struct filedesc *,
-                       sizeof(*newfdp), M_FILEDESC, M_WAITOK);
-       if (newfdp == NULL)
-               return(NULL);
+       newfdp = zalloc(fdp_zone);
  
         proc_fdlock(p);
  
  
         proc_fdlock(p);
  
@@ -4833,8 +4770,9 @@ fdcopy(proc_t p, vnode_t uth_cdir)
          * inherit the new current working directory from the current thread
          * instead, before we take our references.
          */
          * inherit the new current working directory from the current thread
          * instead, before we take our references.
          */
-       if (uth_cdir != NULLVP)
+       if (uth_cdir != NULLVP) {
                 newfdp->fd_cdir = uth_cdir;
                 newfdp->fd_cdir = uth_cdir;
+       }
  
         /*
          * For both fd_cdir and fd_rdir make sure we get
  
         /*
          * For both fd_cdir and fd_rdir make sure we get
@@ -4844,16 +4782,18 @@ fdcopy(proc_t p, vnode_t uth_cdir)
          * and allows us to do the vnode_rele only on
          * a properly referenced vp
          */
          * and allows us to do the vnode_rele only on
          * a properly referenced vp
          */
-       if ( (v_dir = newfdp->fd_cdir) ) {
-               if (vnode_getwithref(v_dir) == 0) {
-                       if ( (vnode_ref(v_dir)) )
-                               newfdp->fd_cdir = NULL;
+       if ((v_dir = newfdp->fd_cdir)) {
+               if (vnode_getwithref(v_dir) == 0) {
+                       if ((vnode_ref(v_dir))) {
+                               newfdp->fd_cdir = NULL;
+                       }
                         vnode_put(v_dir);
                         vnode_put(v_dir);
-               } else
-                       newfdp->fd_cdir = NULL;
+               } else {
+                       newfdp->fd_cdir = NULL;
+               }
         }
         if (newfdp->fd_cdir == NULL && fdp->fd_cdir) {
         }
         if (newfdp->fd_cdir == NULL && fdp->fd_cdir) {
-               /*
+               /*
                  * we couldn't get a new reference on
                  * the current working directory being
                  * inherited... we might as well drop
                  * we couldn't get a new reference on
                  * the current working directory being
                  * inherited... we might as well drop
@@ -4862,22 +4802,24 @@ fdcopy(proc_t p, vnode_t uth_cdir)
                  * it useless... by dropping it we'll
                  * be that much closer to recycling it
                  */
                  * it useless... by dropping it we'll
                  * be that much closer to recycling it
                  */
-               vnode_rele(fdp->fd_cdir);
+               vnode_rele(fdp->fd_cdir);
                 fdp->fd_cdir = NULL;
         }
  
                 fdp->fd_cdir = NULL;
         }
  
-       if ( (v_dir = newfdp->fd_rdir) ) {
+       if ((v_dir = newfdp->fd_rdir)) {
                 if (vnode_getwithref(v_dir) == 0) {
                 if (vnode_getwithref(v_dir) == 0) {
-                       if ( (vnode_ref(v_dir)) )
-                               newfdp->fd_rdir = NULL;
+                       if ((vnode_ref(v_dir))) {
+                               newfdp->fd_rdir = NULL;
+                       }
                         vnode_put(v_dir);
                 } else {
                         vnode_put(v_dir);
                 } else {
-                       newfdp->fd_rdir = NULL;
+                       newfdp->fd_rdir = NULL;
                 }
         }
         /* Coming from a chroot environment and unable to get a reference... */
         if (newfdp->fd_rdir == NULL && fdp->fd_rdir) {
                 }
         }
         /* Coming from a chroot environment and unable to get a reference... */
         if (newfdp->fd_rdir == NULL && fdp->fd_rdir) {
-               /*
+               proc_fdunlock(p);
+               /*
                  * We couldn't get a new reference on
                  * the chroot directory being
                  * inherited... this is fatal, since
                  * We couldn't get a new reference on
                  * the chroot directory being
                  * inherited... this is fatal, since
@@ -4885,10 +4827,11 @@ fdcopy(proc_t p, vnode_t uth_cdir)
                  * escape from a chroot environment by
                  * the new process.
                  */
                  * escape from a chroot environment by
                  * the new process.
                  */
-               if (newfdp->fd_cdir)
-                       vnode_rele(newfdp->fd_cdir);
-               FREE_ZONE(newfdp, sizeof *newfdp, M_FILEDESC);
-               return(NULL);
+               if (newfdp->fd_cdir) {
+                       vnode_rele(newfdp->fd_cdir);
+               }
+               zfree(fdp_zone, newfdp);
+               return NULL;
         }
  
         /*
         }
  
         /*
@@ -4897,32 +4840,34 @@ fdcopy(proc_t p, vnode_t uth_cdir)
          * additional memory for the number of descriptors currently
          * in use.
          */
          * additional memory for the number of descriptors currently
          * in use.
          */
-       if (newfdp->fd_lastfile < NDFILE)
+       if (newfdp->fd_lastfile < NDFILE) {
                 i = NDFILE;
                 i = NDFILE;
-       else {
+       } else {
                 /*
                  * Compute the smallest multiple of NDEXTENT needed
                  * for the file descriptors currently in use,
                  * allowing the table to shrink.
                  */
                 i = newfdp->fd_nfiles;
                 /*
                  * Compute the smallest multiple of NDEXTENT needed
                  * for the file descriptors currently in use,
                  * allowing the table to shrink.
                  */
                 i = newfdp->fd_nfiles;
-               while (i > 1 + 2 * NDEXTENT && i > 1 + newfdp->fd_lastfile * 2)
+               while (i > 1 + 2 * NDEXTENT && i > 1 + newfdp->fd_lastfile * 2) {
                         i /= 2;
                         i /= 2;
+               }
         }
         proc_fdunlock(p);
  
         }
         proc_fdunlock(p);
  
-       MALLOC_ZONE(newfdp->fd_ofiles, struct fileproc **,
-                               i * OFILESIZE, M_OFILETABL, M_WAITOK);
+       newfdp->fd_ofiles = kheap_alloc(KM_OFILETABL, i * OFILESIZE,
+           Z_WAITOK | Z_ZERO);
         if (newfdp->fd_ofiles == NULL) {
         if (newfdp->fd_ofiles == NULL) {
-               if (newfdp->fd_cdir)
-                       vnode_rele(newfdp->fd_cdir);
-               if (newfdp->fd_rdir)
+               if (newfdp->fd_cdir) {
+                       vnode_rele(newfdp->fd_cdir);
+               }
+               if (newfdp->fd_rdir) {
                         vnode_rele(newfdp->fd_rdir);
                         vnode_rele(newfdp->fd_rdir);
+               }
  
  
-               FREE_ZONE(newfdp, sizeof(*newfdp), M_FILEDESC);
-               return(NULL);
+               zfree(fdp_zone, newfdp);
+               return NULL;
         }
         }
-       (void) memset(newfdp->fd_ofiles, 0, i * OFILESIZE);
         proc_fdlock(p);
  
         newfdp->fd_ofileflags = (char *) &newfdp->fd_ofiles[i];
         proc_fdlock(p);
  
         newfdp->fd_ofileflags = (char *) &newfdp->fd_ofiles[i];
@@ -4933,40 +4878,20 @@ fdcopy(proc_t p, vnode_t uth_cdir)
                 char *flags;
  
                 (void) memcpy(newfdp->fd_ofiles, fdp->fd_ofiles,
                 char *flags;
  
                 (void) memcpy(newfdp->fd_ofiles, fdp->fd_ofiles,
-                                       (newfdp->fd_lastfile + 1) * sizeof(*fdp->fd_ofiles));
+                   (newfdp->fd_lastfile + 1) * sizeof(*fdp->fd_ofiles));
                 (void) memcpy(newfdp->fd_ofileflags, fdp->fd_ofileflags,
                 (void) memcpy(newfdp->fd_ofileflags, fdp->fd_ofileflags,
-                                       (newfdp->fd_lastfile + 1) * sizeof(*fdp->fd_ofileflags));
-
-               /*
-                * kq descriptors cannot be copied.
-                */
-               if (newfdp->fd_knlistsize != -1) {
-                       fpp = &newfdp->fd_ofiles[newfdp->fd_lastfile];
-                       flags = &newfdp->fd_ofileflags[newfdp->fd_lastfile];
-                       for (i = newfdp->fd_lastfile;
-                           i >= 0; i--, fpp--, flags--) {
-                               if (*flags & UF_RESERVED)
-                                       continue;       /* (removed below) */
-                               if (*fpp != NULL && (*fpp)->f_type == DTYPE_KQUEUE) {
-                                       *fpp = NULL;
-                                       *flags = 0;
-                                       if (i < newfdp->fd_freefile)
-                                               newfdp->fd_freefile = i;
-                               }
-                               if (*fpp == NULL && i == newfdp->fd_lastfile && i > 0)
-                                       newfdp->fd_lastfile--;
-                       }
-               }
-               fpp = newfdp->fd_ofiles;
-               flags = newfdp->fd_ofileflags;
+                   (newfdp->fd_lastfile + 1) * sizeof(*fdp->fd_ofileflags));
  
  
-               for (i = newfdp->fd_lastfile + 1; --i >= 0; fpp++, flags++)
+               fpp = &newfdp->fd_ofiles[newfdp->fd_lastfile];
+               flags = &newfdp->fd_ofileflags[newfdp->fd_lastfile];
+               for (i = newfdp->fd_lastfile; i >= 0; i--, fpp--, flags--) {
                         if ((ofp = *fpp) != NULL &&
                         if ((ofp = *fpp) != NULL &&
-                           0 == (ofp->f_fglob->fg_lflags & FG_CONFINED) &&
-                           0 == (*flags & (UF_FORKCLOSE|UF_RESERVED))) {
+                           0 == (ofp->fp_glob->fg_lflags & FG_CONFINED) &&
+                           0 == (*flags & (UF_FORKCLOSE | UF_RESERVED))) {
  #if DEBUG
  #if DEBUG
-                               if (FILEPROC_TYPE(ofp) != FTYPE_SIMPLE)
+                               if (FILEPROC_TYPE(ofp) != FTYPE_SIMPLE) {
                                         panic("complex fileproc");
                                         panic("complex fileproc");
+                               }
  #endif
                                 fp = fileproc_alloc_init(NULL);
                                 if (fp == NULL) {
  #endif
                                 fp = fileproc_alloc_init(NULL);
                                 if (fp == NULL) {
@@ -4976,18 +4901,25 @@ fdcopy(proc_t p, vnode_t uth_cdir)
                                          */
                                         *fpp = NULL;
                                 } else {
                                          */
                                         *fpp = NULL;
                                 } else {
-                                       fp->f_flags |=
-                                           (ofp->f_flags & ~FP_TYPEMASK);
-                                       fp->f_fglob = ofp->f_fglob;
-                                       (void)fg_ref(fp);
+                                       fp->fp_flags |=
+                                           (ofp->fp_flags & ~FP_TYPEMASK);
+                                       fp->fp_glob = ofp->fp_glob;
+                                       fg_ref(p, fp->fp_glob);
                                         *fpp = fp;
                                 }
                         } else {
                                         *fpp = fp;
                                 }
                         } else {
-                               if (i < newfdp->fd_freefile)
-                                       newfdp->fd_freefile = i;
                                 *fpp = NULL;
                                 *flags = 0;
                         }
                                 *fpp = NULL;
                                 *flags = 0;
                         }
+                       if (*fpp == NULL) {
+                               if (i == newfdp->fd_lastfile && i > 0) {
+                                       newfdp->fd_lastfile--;
+                               }
+                               if (i < newfdp->fd_freefile) {
+                                       newfdp->fd_freefile = i;
+                               }
+                       }
+               }
         }
  
         proc_fdunlock(p);
         }
  
         proc_fdunlock(p);
@@ -4996,16 +4928,16 @@ fdcopy(proc_t p, vnode_t uth_cdir)
          * Initialize knote and kqueue tracking structs
          */
         newfdp->fd_knlist = NULL;
          * Initialize knote and kqueue tracking structs
          */
         newfdp->fd_knlist = NULL;
-       newfdp->fd_knlistsize = -1;
+       newfdp->fd_knlistsize = 0;
         newfdp->fd_knhash = NULL;
         newfdp->fd_knhashmask = 0;
         newfdp->fd_kqhash = NULL;
         newfdp->fd_kqhashmask = 0;
         newfdp->fd_wqkqueue = NULL;
         newfdp->fd_knhash = NULL;
         newfdp->fd_knhashmask = 0;
         newfdp->fd_kqhash = NULL;
         newfdp->fd_kqhashmask = 0;
         newfdp->fd_wqkqueue = NULL;
-       lck_mtx_init(&newfdp->fd_kqhashlock, proc_kqhashlock_grp, proc_lck_attr);
-       lck_mtx_init(&newfdp->fd_knhashlock, proc_knhashlock_grp, proc_lck_attr);
+       lck_mtx_init(&newfdp->fd_kqhashlock, &proc_kqhashlock_grp, &proc_lck_attr);
+       lck_mtx_init(&newfdp->fd_knhashlock, &proc_knhashlock_grp, &proc_lck_attr);
  
  
-       return (newfdp);
+       return newfdp;
  }
  
  
  }
  
  
@@ -5028,27 +4960,28 @@ fdfree(proc_t p)
  {
         struct filedesc *fdp;
         struct fileproc *fp;
  {
         struct filedesc *fdp;
         struct fileproc *fp;
-       struct kqueue *dealloc_kq = NULL;
+       struct kqworkq *dealloc_kqwq = NULL;
         int i;
  
         proc_fdlock(p);
  
         if (p == kernproc || NULL == (fdp = p->p_fd)) {
         int i;
  
         proc_fdlock(p);
  
         if (p == kernproc || NULL == (fdp = p->p_fd)) {
-               proc_fdunlock(p);
+               proc_fdunlock(p);
                 return;
         }
  
         extern struct filedesc filedesc0;
  
                 return;
         }
  
         extern struct filedesc filedesc0;
  
-       if (&filedesc0 == fdp)
+       if (&filedesc0 == fdp) {
                 panic("filedesc0");
                 panic("filedesc0");
+       }
  
  
-       /* 
+       /*
          * deallocate all the knotes up front and claim empty
          * tables to make any subsequent kqueue closes faster.
          */
         knotes_dealloc(p);
          * deallocate all the knotes up front and claim empty
          * tables to make any subsequent kqueue closes faster.
          */
         knotes_dealloc(p);
-       assert(fdp->fd_knlistsize == -1);
+       assert(fdp->fd_knlistsize == 0);
         assert(fdp->fd_knhashmask == 0);
  
         /*
         assert(fdp->fd_knhashmask == 0);
  
         /*
@@ -5061,151 +4994,52 @@ fdfree(proc_t p)
         if (fdp->fd_nfiles > 0 && fdp->fd_ofiles) {
                 for (i = fdp->fd_lastfile; i >= 0; i--) {
                         if ((fp = fdp->fd_ofiles[i]) != NULL) {
         if (fdp->fd_nfiles > 0 && fdp->fd_ofiles) {
                 for (i = fdp->fd_lastfile; i >= 0; i--) {
                         if ((fp = fdp->fd_ofiles[i]) != NULL) {
-
-                         if (fdp->fd_ofileflags[i] & UF_RESERVED)
-                               panic("fdfree: found fp with UF_RESERVED");
-
-                               procfdtbl_reservefd(p, i);
-
-                               if (fp->f_flags & FP_WAITEVENT)
-                                       (void)waitevent_close(p, fp);
-                               (void) closef_locked(fp, fp->f_fglob, p);
-                               fileproc_free(fp);
+                               if (fdp->fd_ofileflags[i] & UF_RESERVED) {
+                                       panic("fdfree: found fp with UF_RESERVED");
+                               }
+                               fp_close_and_unlock(p, i, fp, 0);
+                               proc_fdlock(p);
                         }
                 }
                         }
                 }
-               FREE_ZONE(fdp->fd_ofiles, fdp->fd_nfiles * OFILESIZE, M_OFILETABL);
+               kheap_free(KM_OFILETABL, fdp->fd_ofiles, fdp->fd_nfiles * OFILESIZE);
                 fdp->fd_ofiles = NULL;
                 fdp->fd_nfiles = 0;
         }
  
         if (fdp->fd_wqkqueue) {
                 fdp->fd_ofiles = NULL;
                 fdp->fd_nfiles = 0;
         }
  
         if (fdp->fd_wqkqueue) {
-               dealloc_kq = fdp->fd_wqkqueue;
-               fdp->fd_wqkqueue = NULL;
-       }
-
-       proc_fdunlock(p);
-
-       if (dealloc_kq)
-               kqueue_dealloc(dealloc_kq);
-
-       if (fdp->fd_cdir)
-               vnode_rele(fdp->fd_cdir);
-       if (fdp->fd_rdir)
-               vnode_rele(fdp->fd_rdir);
-
-       proc_fdlock_spin(p);
-       p->p_fd = NULL;
-       proc_fdunlock(p);
-
-       if (fdp->fd_kqhash) {
-               for (uint32_t j = 0; j <= fdp->fd_kqhashmask; j++)
-                       assert(SLIST_EMPTY(&fdp->fd_kqhash[j]));
-               FREE(fdp->fd_kqhash, M_KQUEUE);
-       }
-
-       lck_mtx_destroy(&fdp->fd_kqhashlock, proc_kqhashlock_grp);
-       lck_mtx_destroy(&fdp->fd_knhashlock, proc_knhashlock_grp);
-
-       FREE_ZONE(fdp, sizeof(*fdp), M_FILEDESC);
-}
-
-/*
- * closef_locked
- *
- * Description:        Internal form of closef; called with proc_fdlock held
- *
- * Parameters: fp                      Pointer to fileproc for fd
- *             fg                      Pointer to fileglob for fd
- *             p                       Pointer to proc structure
- *
- * Returns:    0                       Success
- *     closef_finish:???               Anything returnable by a per-fileops
- *                                     close function
- *
- * Note:       Decrements reference count on file structure; if this was the
- *             last reference, then closef_finish() is called
- *
- *             p and fp are allowed to  be NULL when closing a file that was
- *             being passed in a message (but only if we are called when this
- *             is NOT the last reference).
- */
-int
-closef_locked(struct fileproc *fp, struct fileglob *fg, proc_t p)
-{
-       struct vnode *vp;
-       struct flock lf;
-       struct vfs_context context;
-       int error;
-
-       if (fg == NULL) {
-               return (0);
-       }
-
-       /* Set up context with cred stashed in fg */
-       if (p == current_proc())
-               context.vc_thread = current_thread();
-       else
-               context.vc_thread = NULL;
-       context.vc_ucred = fg->fg_cred;
-
-       /*
-        * POSIX record locking dictates that any close releases ALL
-        * locks owned by this process.  This is handled by setting
-        * a flag in the unlock to free ONLY locks obeying POSIX
-        * semantics, and not to free BSD-style file locks.
-        * If the descriptor was in a message, POSIX-style locks
-        * aren't passed with the descriptor.
-        */
-       if (p && (p->p_ladvflag & P_LADVLOCK) &&
-           DTYPE_VNODE == FILEGLOB_DTYPE(fg)) {
-               proc_fdunlock(p);
-
-               lf.l_whence = SEEK_SET;
-               lf.l_start = 0;
-               lf.l_len = 0;
-               lf.l_type = F_UNLCK;
-               vp = (struct vnode *)fg->fg_data;
-
-               if ( (error = vnode_getwithref(vp)) == 0 ) {
-                       (void) VNOP_ADVLOCK(vp, (caddr_t)p, F_UNLCK, &lf, F_POSIX, &context, NULL);
-                       (void)vnode_put(vp);
-               }
-               proc_fdlock(p);
-       }
-       lck_mtx_lock_spin(&fg->fg_lock);
-       fg->fg_count--;
-
-       if (fg->fg_count > 0) {
-               lck_mtx_unlock(&fg->fg_lock);
-               return (0);
-       }
-#if DIAGNOSTIC
-       if (fg->fg_count != 0)
-               panic("fg %p: being freed with bad fg_count (%d)", fg, fg->fg_count);
-#endif
-
-       if (fp && (fp->f_flags & FP_WRITTEN))
-               fg->fg_flag |= FWASWRITTEN;
+               dealloc_kqwq = fdp->fd_wqkqueue;
+               fdp->fd_wqkqueue = NULL;
+       }
  
  
-       fg->fg_lflags |= FG_TERM;
-       lck_mtx_unlock(&fg->fg_lock);
+       proc_fdunlock(p);
  
  
-       if (p)
-               proc_fdunlock(p);
+       if (dealloc_kqwq) {
+               kqworkq_dealloc(dealloc_kqwq);
+       }
+       if (fdp->fd_cdir) {
+               vnode_rele(fdp->fd_cdir);
+       }
+       if (fdp->fd_rdir) {
+               vnode_rele(fdp->fd_rdir);
+       }
  
  
-       /* Since we ensure that fg->fg_ops is always initialized,
-        * it is safe to invoke fo_close on the fg */
-       error = fo_close(fg, &context);
+       proc_fdlock_spin(p);
+       p->p_fd = NULL;
+       proc_fdunlock(p);
  
  
-       fg_free(fg);
+       if (fdp->fd_kqhash) {
+               for (uint32_t j = 0; j <= fdp->fd_kqhashmask; j++) {
+                       assert(LIST_EMPTY(&fdp->fd_kqhash[j]));
+               }
+               hashdestroy(fdp->fd_kqhash, M_KQUEUE, fdp->fd_kqhashmask);
+       }
  
  
-       if (p)
-               proc_fdlock(p);
+       lck_mtx_destroy(&fdp->fd_kqhashlock, &proc_kqhashlock_grp);
+       lck_mtx_destroy(&fdp->fd_knhashlock, &proc_knhashlock_grp);
  
  
-       return(error);
+       zfree(fdp_zone, fdp);
  }
  
  }
  
-
  /*
   * fileproc_drain
   *
  /*
   * fileproc_drain
   *
@@ -5224,47 +5058,65 @@ closef_locked(struct fileproc *fp, struct fileglob *fg, proc_t p)
   *             close to prevent fd's from being closed out from under
   *             operations currently in progress and blocked
   *
   *             close to prevent fd's from being closed out from under
   *             operations currently in progress and blocked
   *
- * See Also:   file_vnode(), file_socket(), file_drop(), and the cautions
+ * See Also:    file_vnode(), file_socket(), file_drop(), and the cautions
   *             regarding their use and interaction with this function.
   */
  void
  fileproc_drain(proc_t p, struct fileproc * fp)
  {
         struct vfs_context context;
   *             regarding their use and interaction with this function.
   */
  void
  fileproc_drain(proc_t p, struct fileproc * fp)
  {
         struct vfs_context context;
+       thread_t thread;
+       bool is_current_proc;
  
  
-       context.vc_thread = proc_thread(p);     /* XXX */
-       context.vc_ucred = fp->f_fglob->fg_cred;
+       is_current_proc = (p == current_proc());
+
+       if (!is_current_proc) {
+               proc_lock(p);
+               thread = proc_thread(p); /* XXX */
+               thread_reference(thread);
+               proc_unlock(p);
+       } else {
+               thread = current_thread();
+       }
  
  
-       fp->f_iocount-- ; /* (the one the close holds) */
+       context.vc_thread = thread;
+       context.vc_ucred = fp->fp_glob->fg_cred;
  
  
-       while (fp->f_iocount) {
+       /* Set the vflag for drain */
+       fileproc_modify_vflags(fp, FPV_DRAIN, FALSE);
  
  
-               lck_mtx_convert_spin(&p->p_fdmlock);
+       while (os_ref_get_count(&fp->fp_iocount) > 1) {
+               lck_mtx_convert_spin(&p->p_fdmlock);
  
  
-               if (fp->f_fglob->fg_ops->fo_drain) {
-                       (*fp->f_fglob->fg_ops->fo_drain)(fp, &context);
-               }
-               if ((fp->f_flags & FP_INSELECT) == FP_INSELECT) {
-                       if (waitq_wakeup64_all((struct waitq *)fp->f_wset, NO_EVENT64,
-                                              THREAD_INTERRUPTED, WAITQ_ALL_PRIORITIES) == KERN_INVALID_ARGUMENT)
-                               panic("bad wait queue for waitq_wakeup64_all %p (fp:%p)", fp->f_wset, fp);
+               fo_drain(fp, &context);
+               if ((fp->fp_flags & FP_INSELECT) == FP_INSELECT) {
+                       if (waitq_wakeup64_all((struct waitq *)fp->fp_wset, NO_EVENT64,
+                           THREAD_INTERRUPTED, WAITQ_ALL_PRIORITIES) == KERN_INVALID_ARGUMENT) {
+                               panic("bad wait queue for waitq_wakeup64_all %p (fp:%p)", fp->fp_wset, fp);
+                       }
                 }
                 }
-               if ((fp->f_flags & FP_SELCONFLICT) == FP_SELCONFLICT) {
+               if ((fp->fp_flags & FP_SELCONFLICT) == FP_SELCONFLICT) {
                         if (waitq_wakeup64_all(&select_conflict_queue, NO_EVENT64,
                         if (waitq_wakeup64_all(&select_conflict_queue, NO_EVENT64,
-                                              THREAD_INTERRUPTED, WAITQ_ALL_PRIORITIES) == KERN_INVALID_ARGUMENT)
+                           THREAD_INTERRUPTED, WAITQ_ALL_PRIORITIES) == KERN_INVALID_ARGUMENT) {
                                 panic("bad select_conflict_queue");
                                 panic("bad select_conflict_queue");
+                       }
                 }
                 p->p_fpdrainwait = 1;
  
                 msleep(&p->p_fpdrainwait, &p->p_fdmlock, PRIBIO, "fpdrain", NULL);
                 }
                 p->p_fpdrainwait = 1;
  
                 msleep(&p->p_fpdrainwait, &p->p_fdmlock, PRIBIO, "fpdrain", NULL);
-
         }
  #if DIAGNOSTIC
         }
  #if DIAGNOSTIC
-       if ((fp->f_flags & FP_INSELECT) != 0)
+       if ((fp->fp_flags & FP_INSELECT) != 0) {
                 panic("FP_INSELECT set on drained fp");
                 panic("FP_INSELECT set on drained fp");
+       }
  #endif
  #endif
-       if ((fp->f_flags & FP_SELCONFLICT) == FP_SELCONFLICT)
-               fp->f_flags &= ~FP_SELCONFLICT;
+       if ((fp->fp_flags & FP_SELCONFLICT) == FP_SELCONFLICT) {
+               fp->fp_flags &= ~FP_SELCONFLICT;
+       }
+
+       if (!is_current_proc) {
+               thread_deallocate(thread);
+       }
  }
  
  
  }
  
  
@@ -5278,27 +5130,22 @@ fileproc_drain(proc_t p, struct fileproc * fp)
   * Parameters: p                               Process containing fd
   *             fd                              fd to be released
   *             fp                              fileproc to be freed
   * Parameters: p                               Process containing fd
   *             fd                              fd to be released
   *             fp                              fileproc to be freed
- *
- * Returns:    0                               Success
- *
- * Notes:      XXX function should be void - no one interprets the returns
- *             XXX code
   */
   */
-int
+void
  fp_free(proc_t p, int fd, struct fileproc * fp)
  {
  fp_free(proc_t p, int fd, struct fileproc * fp)
  {
-        proc_fdlock_spin(p);
+       proc_fdlock_spin(p);
         fdrelse(p, fd);
         fdrelse(p, fd);
-        proc_fdunlock(p);
+       proc_fdunlock(p);
  
  
-       fg_free(fp->f_fglob);
+       fg_free(fp->fp_glob);
+       os_ref_release_live(&fp->fp_iocount);
         fileproc_free(fp);
         fileproc_free(fp);
-       return(0);
  }
  
  
  /*
  }
  
  
  /*
- * flock
+ * sys_flock
   *
   * Description:        Apply an advisory lock on a file descriptor.
   *
   *
   * Description:        Apply an advisory lock on a file descriptor.
   *
@@ -5321,7 +5168,7 @@ fp_free(proc_t p, int fd, struct fileproc * fp)
   *             the entire file (l_whence = SEEK_SET, l_start = 0, l_len = 0).
   */
  int
   *             the entire file (l_whence = SEEK_SET, l_start = 0, l_len = 0).
   */
  int
-flock(proc_t p, struct flock_args *uap, __unused int32_t *retval)
+sys_flock(proc_t p, struct flock_args *uap, __unused int32_t *retval)
  {
         int fd = uap->fd;
         int how = uap->how;
  {
         int fd = uap->fd;
         int how = uap->how;
@@ -5329,13 +5176,13 @@ flock(proc_t p, struct flock_args *uap, __unused int32_t *retval)
         struct vnode *vp;
         struct flock lf;
         vfs_context_t ctx = vfs_context_current();
         struct vnode *vp;
         struct flock lf;
         vfs_context_t ctx = vfs_context_current();
-       int error=0;
+       int error = 0;
  
         AUDIT_ARG(fd, uap->fd);
  
         AUDIT_ARG(fd, uap->fd);
-       if ( (error = fp_getfvp(p, fd, &fp, &vp)) ) {
-               return(error);
+       if ((error = fp_getfvp(p, fd, &fp, &vp))) {
+               return error;
         }
         }
-       if ( (error = vnode_getwithref(vp)) ) {
+       if ((error = vnode_getwithref(vp))) {
                 goto out1;
         }
         AUDIT_ARG(vnpath, vp, ARG_VNODE1);
                 goto out1;
         }
         AUDIT_ARG(vnpath, vp, ARG_VNODE1);
@@ -5345,56 +5192,56 @@ flock(proc_t p, struct flock_args *uap, __unused int32_t *retval)
         lf.l_len = 0;
         if (how & LOCK_UN) {
                 lf.l_type = F_UNLCK;
         lf.l_len = 0;
         if (how & LOCK_UN) {
                 lf.l_type = F_UNLCK;
-               fp->f_flag &= ~FHASLOCK;
-               error = VNOP_ADVLOCK(vp, (caddr_t)fp->f_fglob, F_UNLCK, &lf, F_FLOCK, ctx, NULL);
+               error = VNOP_ADVLOCK(vp, (caddr_t)fp->fp_glob, F_UNLCK, &lf, F_FLOCK, ctx, NULL);
                 goto out;
         }
                 goto out;
         }
-       if (how & LOCK_EX)
+       if (how & LOCK_EX) {
                 lf.l_type = F_WRLCK;
                 lf.l_type = F_WRLCK;
-       else if (how & LOCK_SH)
+       } else if (how & LOCK_SH) {
                 lf.l_type = F_RDLCK;
                 lf.l_type = F_RDLCK;
-       else {
-               error = EBADF;
+       } else {
+               error = EBADF;
                 goto out;
         }
  #if CONFIG_MACF
                 goto out;
         }
  #if CONFIG_MACF
-       error = mac_file_check_lock(proc_ucred(p), fp->f_fglob, F_SETLK, &lf);
-       if (error)
+       error = mac_file_check_lock(proc_ucred(p), fp->fp_glob, F_SETLK, &lf);
+       if (error) {
                 goto out;
                 goto out;
+       }
  #endif
  #endif
-       error = VNOP_ADVLOCK(vp, (caddr_t)fp->f_fglob, F_SETLK, &lf,
+       error = VNOP_ADVLOCK(vp, (caddr_t)fp->fp_glob, F_SETLK, &lf,
             (how & LOCK_NB ? F_FLOCK : F_FLOCK | F_WAIT),
             ctx, NULL);
             (how & LOCK_NB ? F_FLOCK : F_FLOCK | F_WAIT),
             ctx, NULL);
-       if (!error)
-               fp->f_flag |= FHASLOCK;
+       if (!error) {
+               os_atomic_or(&fp->fp_glob->fg_flag, FWASLOCKED, relaxed);
+       }
  out:
         (void)vnode_put(vp);
  out1:
         fp_drop(p, fd, fp, 0);
  out:
         (void)vnode_put(vp);
  out1:
         fp_drop(p, fd, fp, 0);
-       return(error);
-
+       return error;
  }
  
  /*
  }
  
  /*
- * fileport_makeport
+ * sys_fileport_makeport
   *
   * Description: Obtain a Mach send right for a given file descriptor.
   *
   * Parameters: p               Process calling fileport
   *
   * Description: Obtain a Mach send right for a given file descriptor.
   *
   * Parameters: p               Process calling fileport
- *             uap->fd         The fd to reference
- *             uap->portnamep  User address at which to place port name.
+ *              uap->fd                The fd to reference
+ *              uap->portnamep  User address at which to place port name.
   *
   * Returns:    0               Success.
   *
   * Returns:    0               Success.
- *                     EBADF           Bad file descriptor.
- *                     EINVAL          File descriptor had type that cannot be sent, misc. other errors.
- *                     EFAULT          Address at which to store port name is not valid.
- *                     EAGAIN          Resource shortage.
+ *              EBADF          Bad file descriptor.
+ *              EINVAL         File descriptor had type that cannot be sent, misc. other errors.
+ *              EFAULT         Address at which to store port name is not valid.
+ *              EAGAIN         Resource shortage.
   *
   * Implicit returns:
   *             On success, name of send right is stored at user-specified address.
   */
  int
   *
   * Implicit returns:
   *             On success, name of send right is stored at user-specified address.
   */
  int
-fileport_makeport(proc_t p, struct fileport_makeport_args *uap,
+sys_fileport_makeport(proc_t p, struct fileport_makeport_args *uap,
      __unused int *retval)
  {
         int err;
      __unused int *retval)
  {
         int err;
@@ -5411,27 +5258,27 @@ fileport_makeport(proc_t p, struct fileport_makeport_args *uap,
                 goto out_unlock;
         }
  
                 goto out_unlock;
         }
  
-       if (!file_issendable(p, fp)) {
+       fg = fp->fp_glob;
+       if (!fg_sendable(fg)) {
                 err = EINVAL;
                 goto out_unlock;
         }
  
                 err = EINVAL;
                 goto out_unlock;
         }
  
-       if (FP_ISGUARDED(fp, GUARD_FILEPORT)) {
+       if (fp_isguarded(fp, GUARD_FILEPORT)) {
                 err = fp_guard_exception(p, fd, fp, kGUARD_EXC_FILEPORT);
                 goto out_unlock;
         }
  
         /* Dropped when port is deallocated */
                 err = fp_guard_exception(p, fd, fp, kGUARD_EXC_FILEPORT);
                 goto out_unlock;
         }
  
         /* Dropped when port is deallocated */
-       fg = fp->f_fglob;
-       fg_ref(fp);
+       fg_ref(p, fg);
  
         proc_fdunlock(p);
  
         /* Allocate and initialize a port */
         fileport = fileport_alloc(fg);
         if (fileport == IPC_PORT_NULL) {
  
         proc_fdunlock(p);
  
         /* Allocate and initialize a port */
         fileport = fileport_alloc(fg);
         if (fileport == IPC_PORT_NULL) {
+               fg_drop_live(fg);
                 err = EAGAIN;
                 err = EAGAIN;
-               fg_drop(fp);
                 goto out;
         }
  
                 goto out;
         }
  
@@ -5474,20 +5321,14 @@ out:
  void
  fileport_releasefg(struct fileglob *fg)
  {
  void
  fileport_releasefg(struct fileglob *fg)
  {
-       (void)closef_locked(NULL, fg, PROC_NULL);
-
-       return;
+       (void)fg_drop(PROC_NULL, fg);
  }
  
  }
  
-
  /*
   * fileport_makefd
   *
   * Description: Obtain the file descriptor for a given Mach send right.
   *
  /*
   * fileport_makefd
   *
   * Description: Obtain the file descriptor for a given Mach send right.
   *
- * Parameters: p               Process calling fileport
- *             uap->port       Name of send right to file port.
- *
   * Returns:    0               Success
   *             EINVAL          Invalid Mach port name, or port is not for a file.
   *     fdalloc:EMFILE
   * Returns:    0               Success
   *             EINVAL          Invalid Mach port name, or port is not for a file.
   *     fdalloc:EMFILE
@@ -5497,24 +5338,13 @@ fileport_releasefg(struct fileglob *fg)
   *             *retval (modified)              The new descriptor
   */
  int
   *             *retval (modified)              The new descriptor
   */
  int
-fileport_makefd(proc_t p, struct fileport_makefd_args *uap, int32_t *retval)
+fileport_makefd(proc_t p, ipc_port_t port, int uf_flags, int *retval)
  {
         struct fileglob *fg;
  {
         struct fileglob *fg;
-       struct fileproc *fp = FILEPROC_NULL;
-       ipc_port_t port = IPC_PORT_NULL;
-       mach_port_name_t send = uap->port;
-       kern_return_t res;
+       struct fileproc *fp = FILEPROC_NULL;
         int fd;
         int err;
  
         int fd;
         int err;
  
-       res = ipc_object_copyin(get_task_ipcspace(p->task),
-                       send, MACH_MSG_TYPE_COPY_SEND, &port);
-
-       if (res != KERN_SUCCESS) {
-               err = EINVAL;
-               goto out;
-       }
-
         fg = fileport_port_to_fileglob(port);
         if (fg == NULL) {
                 err = EINVAL;
         fg = fileport_port_to_fileglob(port);
         if (fg == NULL) {
                 err = EINVAL;
@@ -5527,17 +5357,18 @@ fileport_makefd(proc_t p, struct fileport_makefd_args *uap, int32_t *retval)
                 goto out;
         }
  
                 goto out;
         }
  
-       fp->f_fglob = fg;
-       fg_ref(fp);
-
-       proc_fdlock(p);
+       proc_fdlock(p);
         err = fdalloc(p, 0, &fd);
         if (err != 0) {
                 proc_fdunlock(p);
         err = fdalloc(p, 0, &fd);
         if (err != 0) {
                 proc_fdunlock(p);
-               fg_drop(fp);
                 goto out;
         }
                 goto out;
         }
-       *fdflags(p, fd) |= UF_EXCLOSE;
+       if (uf_flags) {
+               *fdflags(p, fd) |= uf_flags;
+       }
+
+       fp->fp_glob = fg;
+       fg_ref(p, fg);
  
         procfdtbl_releasefd(p, fd, fp);
         proc_fdunlock(p);
  
         procfdtbl_releasefd(p, fd, fp);
         proc_fdunlock(p);
@@ -5549,6 +5380,42 @@ out:
                 fileproc_free(fp);
         }
  
                 fileproc_free(fp);
         }
  
+       return err;
+}
+
+/*
+ * sys_fileport_makefd
+ *
+ * Description: Obtain the file descriptor for a given Mach send right.
+ *
+ * Parameters: p               Process calling fileport
+ *              uap->port      Name of send right to file port.
+ *
+ * Returns:    0               Success
+ *             EINVAL          Invalid Mach port name, or port is not for a file.
+ *     fdalloc:EMFILE
+ *     fdalloc:ENOMEM          Unable to allocate fileproc or extend file table.
+ *
+ * Implicit returns:
+ *             *retval (modified)              The new descriptor
+ */
+int
+sys_fileport_makefd(proc_t p, struct fileport_makefd_args *uap, int32_t *retval)
+{
+       ipc_port_t port = IPC_PORT_NULL;
+       mach_port_name_t send = uap->port;
+       kern_return_t res;
+       int err;
+
+       res = ipc_object_copyin(get_task_ipcspace(p->task),
+           send, MACH_MSG_TYPE_COPY_SEND, &port, 0, NULL, IPC_OBJECT_COPYIN_FLAGS_ALLOW_IMMOVABLE_SEND);
+
+       if (res == KERN_SUCCESS) {
+               err = fileport_makefd(p, port, UF_EXCLOSE, retval);
+       } else {
+               err = EINVAL;
+       }
+
         if (IPC_PORT_NULL != port) {
                 ipc_port_release_send(port);
         }
         if (IPC_PORT_NULL != port) {
                 ipc_port_release_send(port);
         }
@@ -5598,17 +5465,16 @@ dupfdopen(struct filedesc *fdp, int indx, int dfd, int flags, int error)
  
         fp = fdp->fd_ofiles[indx];
         if (dfd < 0 || dfd >= fdp->fd_nfiles ||
  
         fp = fdp->fd_ofiles[indx];
         if (dfd < 0 || dfd >= fdp->fd_nfiles ||
-                       (wfp = fdp->fd_ofiles[dfd]) == NULL || wfp == fp ||
-                       (fdp->fd_ofileflags[dfd] & UF_RESERVED)) {
-
-               proc_fdunlock(p);
-               return (EBADF);
+           (wfp = fdp->fd_ofiles[dfd]) == NULL || wfp == fp ||
+           (fdp->fd_ofileflags[dfd] & UF_RESERVED)) {
+               proc_fdunlock(p);
+               return EBADF;
         }
  #if CONFIG_MACF
         }
  #if CONFIG_MACF
-       myerror = mac_file_check_dup(proc_ucred(p), wfp->f_fglob, dfd);
+       myerror = mac_file_check_dup(proc_ucred(p), wfp->fp_glob, dfd);
         if (myerror) {
                 proc_fdunlock(p);
         if (myerror) {
                 proc_fdunlock(p);
-               return (myerror);
+               return myerror;
         }
  #endif
         /*
         }
  #endif
         /*
@@ -5625,232 +5491,43 @@ dupfdopen(struct filedesc *fdp, int indx, int dfd, int flags, int error)
          */
         switch (error) {
         case ENODEV:
          */
         switch (error) {
         case ENODEV:
-               if (FP_ISGUARDED(wfp, GUARD_DUP)) {
+               if (fp_isguarded(wfp, GUARD_DUP)) {
                         proc_fdunlock(p);
                         proc_fdunlock(p);
-                       return (EPERM);
+                       return EPERM;
                 }
  
                 /*
                  * Check that the mode the file is being opened for is a
                  * subset of the mode of the existing descriptor.
                  */
                 }
  
                 /*
                  * Check that the mode the file is being opened for is a
                  * subset of the mode of the existing descriptor.
                  */
-               if (((flags & (FREAD|FWRITE)) | wfp->f_flag) != wfp->f_flag) {
-                       proc_fdunlock(p);
-                       return (EACCES);
+               if (((flags & (FREAD | FWRITE)) | wfp->f_flag) != wfp->f_flag) {
+                       proc_fdunlock(p);
+                       return EACCES;
                 }
                 }
-               if (indx > fdp->fd_lastfile)
+               if (indx > fdp->fd_lastfile) {
                         fdp->fd_lastfile = indx;
                         fdp->fd_lastfile = indx;
-               (void)fg_ref(wfp);
+               }
  
  
-               if (fp->f_fglob)
-                       fg_free(fp->f_fglob);
-               fp->f_fglob = wfp->f_fglob;
+               if (fp->fp_glob) {
+                       fg_free(fp->fp_glob);
+               }
+               fg_ref(p, wfp->fp_glob);
+               fp->fp_glob = wfp->fp_glob;
  
                 fdp->fd_ofileflags[indx] = fdp->fd_ofileflags[dfd] |
  
                 fdp->fd_ofileflags[indx] = fdp->fd_ofileflags[dfd] |
-                       (flags & O_CLOEXEC) ? UF_EXCLOSE : 0;
+                   (flags & O_CLOEXEC) ? UF_EXCLOSE : 0;
  
  
-               proc_fdunlock(p);
-               return (0);
+               proc_fdunlock(p);
+               return 0;
  
         default:
  
         default:
-               proc_fdunlock(p);
-               return (error);
+               proc_fdunlock(p);
+               return error;
         }
         /* NOTREACHED */
  }
  
  
         }
         /* NOTREACHED */
  }
  
  
-/*
- * fg_ref
- *
- * Description:        Add a reference to a fileglob by fileproc
- *
- * Parameters: fp                              fileproc containing fileglob
- *                                             pointer
- *
- * Returns:    void
- *
- * Notes:      XXX Should use OSAddAtomic?
- */
-void
-fg_ref(struct fileproc * fp)
-{
-       struct fileglob *fg;
-
-       fg = fp->f_fglob;
-
-       lck_mtx_lock_spin(&fg->fg_lock);
-
-#if DIAGNOSTIC
-       if ((fp->f_flags & ~((unsigned int)FP_VALID_FLAGS)) != 0)
-               panic("fg_ref: invalid bits on fp %p", fp);
-
-       if (fg->fg_count == 0)
-               panic("fg_ref: adding fgcount to zeroed fg: fp %p fg %p",
-                   fp, fg);
-#endif
-       fg->fg_count++;
-       lck_mtx_unlock(&fg->fg_lock);
-}
-
-
-/*
- * fg_drop
- *
- * Description:        Remove a reference to a fileglob by fileproc
- *
- * Parameters: fp                              fileproc containing fileglob
- *                                             pointer
- *
- * Returns:    void
- *
- * Notes:      XXX Should use OSAddAtomic?
- */
-void
-fg_drop(struct fileproc * fp)
-{
-       struct fileglob *fg;
-
-       fg = fp->f_fglob;
-       lck_mtx_lock_spin(&fg->fg_lock);
-       fg->fg_count--;
-       lck_mtx_unlock(&fg->fg_lock);
-}
-
-#if SOCKETS
-/*
- * fg_insertuipc_mark
- *
- * Description:        Mark fileglob for insertion onto message queue if needed
- *             Also takes fileglob reference
- *
- * Parameters: fg      Fileglob pointer to insert
- *
- * Returns:    true, if the fileglob needs to be inserted onto msg queue
- *
- * Locks:      Takes and drops fg_lock, potentially many times
- */
-boolean_t
-fg_insertuipc_mark(struct fileglob * fg)
-{
-       boolean_t insert = FALSE;
-
-       lck_mtx_lock_spin(&fg->fg_lock);
-       while (fg->fg_lflags & FG_RMMSGQ) {
-               lck_mtx_convert_spin(&fg->fg_lock);
-
-               fg->fg_lflags |= FG_WRMMSGQ;
-               msleep(&fg->fg_lflags, &fg->fg_lock, 0, "fg_insertuipc", NULL);
-       }
-
-       fg->fg_count++;
-       fg->fg_msgcount++;
-       if (fg->fg_msgcount == 1) {
-               fg->fg_lflags |= FG_INSMSGQ;
-               insert = TRUE;
-       }
-       lck_mtx_unlock(&fg->fg_lock);
-       return (insert);
-}
-
-/*
- * fg_insertuipc
- *
- * Description:        Insert marked fileglob onto message queue
- *
- * Parameters: fg      Fileglob pointer to insert
- *
- * Returns:    void
- *
- * Locks:      Takes and drops fg_lock & uipc_lock
- *             DO NOT call this function with proc_fdlock held as unp_gc()
- *             can potentially try to acquire proc_fdlock, which can result
- *             in a deadlock if this function is in unp_gc_wait().
- */
-void
-fg_insertuipc(struct fileglob * fg)
-{
-       if (fg->fg_lflags & FG_INSMSGQ) {
-               lck_mtx_lock_spin(uipc_lock);
-               unp_gc_wait();
-               LIST_INSERT_HEAD(&fmsghead, fg, f_msglist);
-               lck_mtx_unlock(uipc_lock);
-               lck_mtx_lock(&fg->fg_lock);
-               fg->fg_lflags &= ~FG_INSMSGQ;
-               if (fg->fg_lflags & FG_WINSMSGQ) {
-                       fg->fg_lflags &= ~FG_WINSMSGQ;
-                       wakeup(&fg->fg_lflags);
-               }
-               lck_mtx_unlock(&fg->fg_lock);
-       }
-}
-
-/*
- * fg_removeuipc_mark
- *
- * Description:        Mark the fileglob for removal from message queue if needed
- *             Also releases fileglob message queue reference
- *
- * Parameters: fg      Fileglob pointer to remove
- *
- * Returns:    true, if the fileglob needs to be removed from msg queue
- *
- * Locks:      Takes and drops fg_lock, potentially many times
- */
-boolean_t
-fg_removeuipc_mark(struct fileglob * fg)
-{
-       boolean_t remove = FALSE;
-
-       lck_mtx_lock_spin(&fg->fg_lock);
-       while (fg->fg_lflags & FG_INSMSGQ) {
-               lck_mtx_convert_spin(&fg->fg_lock);
-
-               fg->fg_lflags |= FG_WINSMSGQ;
-               msleep(&fg->fg_lflags, &fg->fg_lock, 0, "fg_removeuipc", NULL);
-       }
-       fg->fg_msgcount--;
-       if (fg->fg_msgcount == 0) {
-               fg->fg_lflags |= FG_RMMSGQ;
-               remove = TRUE;
-       }
-       lck_mtx_unlock(&fg->fg_lock);
-       return (remove);
-}
-
-/*
- * fg_removeuipc
- *
- * Description:        Remove marked fileglob from message queue
- *
- * Parameters: fg      Fileglob pointer to remove
- *
- * Returns:    void
- *
- * Locks:      Takes and drops fg_lock & uipc_lock
- *             DO NOT call this function with proc_fdlock held as unp_gc()
- *             can potentially try to acquire proc_fdlock, which can result
- *             in a deadlock if this function is in unp_gc_wait().
- */
-void
-fg_removeuipc(struct fileglob * fg)
-{
-       if (fg->fg_lflags & FG_RMMSGQ) {
-               lck_mtx_lock_spin(uipc_lock);
-               unp_gc_wait();
-               LIST_REMOVE(fg, f_msglist);
-               lck_mtx_unlock(uipc_lock);
-               lck_mtx_lock(&fg->fg_lock);
-               fg->fg_lflags &= ~FG_RMMSGQ;
-               if (fg->fg_lflags & FG_WRMMSGQ) {
-                       fg->fg_lflags &= ~FG_WRMMSGQ;
-                       wakeup(&fg->fg_lflags);
-               }
-               lck_mtx_unlock(&fg->fg_lock);
-       }
-}
-#endif /* SOCKETS */
-
  /*
   * fo_read
   *
  /*
   * fo_read
   *
@@ -5868,7 +5545,14 @@ fg_removeuipc(struct fileglob * fg)
  int
  fo_read(struct fileproc *fp, struct uio *uio, int flags, vfs_context_t ctx)
  {
  int
  fo_read(struct fileproc *fp, struct uio *uio, int flags, vfs_context_t ctx)
  {
-       return ((*fp->f_ops->fo_read)(fp, uio, flags, ctx));
+       return (*fp->f_ops->fo_read)(fp, uio, flags, ctx);
+}
+
+int
+fo_no_read(struct fileproc *fp, struct uio *uio, int flags, vfs_context_t ctx)
+{
+#pragma unused(fp, uio, flags, ctx)
+       return ENXIO;
  }
  
  
  }
  
  
@@ -5889,7 +5573,14 @@ fo_read(struct fileproc *fp, struct uio *uio, int flags, vfs_context_t ctx)
  int
  fo_write(struct fileproc *fp, struct uio *uio, int flags, vfs_context_t ctx)
  {
  int
  fo_write(struct fileproc *fp, struct uio *uio, int flags, vfs_context_t ctx)
  {
-       return((*fp->f_ops->fo_write)(fp, uio, flags, ctx));
+       return (*fp->f_ops->fo_write)(fp, uio, flags, ctx);
+}
+
+int
+fo_no_write(struct fileproc *fp, struct uio *uio, int flags, vfs_context_t ctx)
+{
+#pragma unused(fp, uio, flags, ctx)
+       return ENXIO;
  }
  
  
  }
  
  
@@ -5923,7 +5614,14 @@ fo_ioctl(struct fileproc *fp, u_long com, caddr_t data, vfs_context_t ctx)
         proc_fdunlock(vfs_context_proc(ctx));
         error = (*fp->f_ops->fo_ioctl)(fp, com, data, ctx);
         proc_fdlock(vfs_context_proc(ctx));
         proc_fdunlock(vfs_context_proc(ctx));
         error = (*fp->f_ops->fo_ioctl)(fp, com, data, ctx);
         proc_fdlock(vfs_context_proc(ctx));
-       return(error);
+       return error;
+}
+
+int
+fo_no_ioctl(struct fileproc *fp, u_long com, caddr_t data, vfs_context_t ctx)
+{
+#pragma unused(fp, com, data, ctx)
+       return ENOTTY;
  }
  
  
  }
  
  
@@ -5944,7 +5642,14 @@ fo_ioctl(struct fileproc *fp, u_long com, caddr_t data, vfs_context_t ctx)
  int
  fo_select(struct fileproc *fp, int which, void *wql, vfs_context_t ctx)
  {
  int
  fo_select(struct fileproc *fp, int which, void *wql, vfs_context_t ctx)
  {
-       return((*fp->f_ops->fo_select)(fp, which, wql, ctx));
+       return (*fp->f_ops->fo_select)(fp, which, wql, ctx);
+}
+
+int
+fo_no_select(struct fileproc *fp, int which, void *wql, vfs_context_t ctx)
+{
+#pragma unused(fp, which, wql, ctx)
+       return ENOTSUP;
  }
  
  
  }
  
  
@@ -5964,7 +5669,33 @@ fo_select(struct fileproc *fp, int which, void *wql, vfs_context_t ctx)
  int
  fo_close(struct fileglob *fg, vfs_context_t ctx)
  {
  int
  fo_close(struct fileglob *fg, vfs_context_t ctx)
  {
-       return((*fg->fg_ops->fo_close)(fg, ctx));
+       return (*fg->fg_ops->fo_close)(fg, ctx);
+}
+
+
+/*
+ * fo_drain
+ *
+ * Description:        Generic fileops kqueue filter indirected through the fileops
+ *             pointer in the fileproc structure
+ *
+ * Parameters: fp                              fileproc structure pointer
+ *             ctx                             VFS context for operation
+ *
+ * Returns:    0                               Success
+ *             !0                              errno from drain
+ */
+int
+fo_drain(struct fileproc *fp, vfs_context_t ctx)
+{
+       return (*fp->f_ops->fo_drain)(fp, ctx);
+}
+
+int
+fo_no_drain(struct fileproc *fp, vfs_context_t ctx)
+{
+#pragma unused(fp, ctx)
+       return ENOTSUP;
  }
  
  
  }
  
  
@@ -5976,65 +5707,70 @@ fo_close(struct fileglob *fg, vfs_context_t ctx)
   *
   * Parameters: fp                              fileproc structure pointer
   *             kn                              pointer to knote to filter on
   *
   * Parameters: fp                              fileproc structure pointer
   *             kn                              pointer to knote to filter on
- *             ctx                             VFS context for operation
   *
   * Returns:    (kn->kn_flags & EV_ERROR)       error in kn->kn_data
   *             0                               Filter is not active
   *             !0                              Filter is active
   */
  int
   *
   * Returns:    (kn->kn_flags & EV_ERROR)       error in kn->kn_data
   *             0                               Filter is not active
   *             !0                              Filter is active
   */
  int
-fo_kqfilter(struct fileproc *fp, struct knote *kn,
-               struct kevent_internal_s *kev, vfs_context_t ctx)
+fo_kqfilter(struct fileproc *fp, struct knote *kn, struct kevent_qos_s *kev)
  {
  {
-        return ((*fp->f_ops->fo_kqfilter)(fp, kn, kev, ctx));
+       return (*fp->f_ops->fo_kqfilter)(fp, kn, kev);
  }
  
  }
  
-/*
- * The ability to send a file descriptor to another
- * process is opt-in by file type.
- */
-boolean_t
-file_issendable(proc_t p, struct fileproc *fp)
+int
+fo_no_kqfilter(struct fileproc *fp, struct knote *kn, struct kevent_qos_s *kev)
  {
  {
-       proc_fdlock_assert(p, LCK_MTX_ASSERT_OWNED);
-
-       switch (fp->f_type) {
-       case DTYPE_VNODE:
-       case DTYPE_SOCKET:
-       case DTYPE_PIPE:
-       case DTYPE_PSXSHM:
-       case DTYPE_NETPOLICY:
-               return (0 == (fp->f_fglob->fg_lflags & FG_CONFINED));
-       default:
-               /* DTYPE_KQUEUE, DTYPE_FSEVENTS, DTYPE_PSXSEM */
-               return FALSE;
-       }
+#pragma unused(fp, kev)
+       knote_set_error(kn, ENOTSUP);
+       return 0;
  }
  
  
  struct fileproc *
  fileproc_alloc_init(__unused void *arg)
  {
  }
  
  
  struct fileproc *
  fileproc_alloc_init(__unused void *arg)
  {
-       struct fileproc *fp;
-
-       MALLOC_ZONE(fp, struct fileproc *, sizeof (*fp), M_FILEPROC, M_WAITOK);
-       if (fp)
-               bzero(fp, sizeof (*fp));
+       struct fileproc *fp = zalloc_flags(fp_zone, Z_WAITOK | Z_ZERO);
  
  
-       return (fp);
+       os_ref_init(&fp->fp_iocount, &f_refgrp);
+       return fp;
  }
  
  }
  
+
  void
  fileproc_free(struct fileproc *fp)
  {
  void
  fileproc_free(struct fileproc *fp)
  {
+       os_ref_count_t __unused refc = os_ref_release(&fp->fp_iocount);
+#if DEVELOPMENT || DEBUG
+       if (0 != refc) {
+               panic("%s: pid %d refc: %u != 0",
+                   __func__, proc_pid(current_proc()), refc);
+       }
+#endif
         switch (FILEPROC_TYPE(fp)) {
         case FTYPE_SIMPLE:
         switch (FILEPROC_TYPE(fp)) {
         case FTYPE_SIMPLE:
-               FREE_ZONE(fp, sizeof (*fp), M_FILEPROC);
+               zfree(fp_zone, fp);
                 break;
         case FTYPE_GUARDED:
                 guarded_fileproc_free(fp);
                 break;
         default:
                 break;
         case FTYPE_GUARDED:
                 guarded_fileproc_free(fp);
                 break;
         default:
-               panic("%s: corrupt fp %p flags %x", __func__, fp, fp->f_flags);
+               panic("%s: corrupt fp %p flags %x", __func__, fp, fp->fp_flags);
+       }
+}
+
+void
+fileproc_modify_vflags(struct fileproc *fp, fileproc_vflags_t vflags, boolean_t clearflags)
+{
+       if (clearflags) {
+               os_atomic_andnot(&fp->fp_vflags, vflags, relaxed);
+       } else {
+               os_atomic_or(&fp->fp_vflags, vflags, relaxed);
         }
  }
         }
  }
+
+fileproc_vflags_t
+fileproc_get_vflags(struct fileproc *fp)
+{
+       return os_atomic_load(&fp->fp_vflags, relaxed);
+}