X-Git-Url: https://git.saurik.com/apple/xnu.git/blobdiff_plain/21362eb3e66fd2c787aee132bce100a44d71a99c..c7d2c2c6ee645e10cbccdd01c6191873ec77239d:/bsd/kern/kern_descrip.c diff --git a/bsd/kern/kern_descrip.c b/bsd/kern/kern_descrip.c index e147a30bf..62a9d94f7 100644 --- a/bsd/kern/kern_descrip.c +++ b/bsd/kern/kern_descrip.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2000-2004 Apple Computer, Inc. All rights reserved. + * Copyright (c) 2000-2015 Apple Inc. All rights reserved. * * @APPLE_OSREFERENCE_LICENSE_HEADER_START@ * @@ -65,6 +65,12 @@ * * @(#)kern_descrip.c 8.8 (Berkeley) 2/14/95 */ +/* + * NOTICE: This file was modified by SPARTA, Inc. in 2006 to introduce + * support for mandatory and extensible security protections. This notice + * is included in support of clause 2.2 (b) of the Apple Public License, + * Version 2.0. + */ #include #include @@ -74,6 +80,8 @@ #include #include #include +#include +#include #include #include #include @@ -86,55 +94,81 @@ #include #include #include -#include +#include +#include +#include +#include -#include +#include #include #include #include #include +#include #include #include +#include #include +#include + +#include +#include + +#include +#include + +#include + +kern_return_t ipc_object_copyin(ipc_space_t, mach_port_name_t, + mach_msg_type_name_t, ipc_port_t *); +void ipc_port_release_send(ipc_port_t); + struct psemnode; struct pshmnode; -int fdopen(dev_t dev, int mode, int type, struct proc *p); -int ogetdtablesize(struct proc *p, void *uap, register_t *retval); -int finishdup(struct proc * p, struct filedesc *fdp, int old, int new, register_t *retval); +static int finishdup(proc_t p, + struct filedesc *fdp, int old, int new, int flags, int32_t *retval); -int closef(struct fileglob *fg, struct proc *p); -int falloc_locked(struct proc *p, struct fileproc **resultfp, int *resultfd, int locked); -void fddrop(struct proc *p, int fd); -int fdgetf_noref(struct proc *p, int fd, struct fileproc **resultfp); +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); -int fp_getfpshm(struct proc *p, int fd, struct fileproc **resultfp, struct pshmnode **resultpshm); +void fileport_releasefg(struct fileglob *fg); + +/* flags for close_internal_locked */ +#define FD_DUP2RESV 1 + +/* We don't want these exported */ + +__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 int closef_finish(struct fileproc *fp, struct fileglob *fg, struct proc *p); extern void file_lock_init(void); -extern int is_suser(void); -extern int kqueue_stat(struct fileproc *fp, struct stat *st, struct proc *p); -extern int soo_stat(struct socket *so, struct stat *ub); -extern int vn_path_package_check(vnode_t vp, char *path, int pathlen, int *component); extern kauth_scope_t kauth_scope_fileop; +/* Conflict wait queue for when selects collide (opaque type) */ +extern struct waitq select_conflict_queue; + #define f_flag f_fglob->fg_flag -#define f_type f_fglob->fg_type +#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 CHECK_ADD_OVERFLOW_INT64L(x, y) \ + (((((x) > 0) && ((y) > 0) && ((x) > LLONG_MAX - (y))) || \ + (((x) < 0) && ((y) < 0) && ((x) < LLONG_MIN - (y)))) \ + ? 1 : 0) /* * Descriptor management. */ -struct filelist filehead; /* head of list of open files */ 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 */ @@ -145,99 +179,349 @@ lck_grp_t * file_lck_grp; lck_attr_t * file_lck_attr; lck_mtx_t * uipc_lock; -lck_mtx_t * file_iterate_lcok; -lck_mtx_t * file_flist_lock; +/* + * check_file_seek_range + * + * Description: Checks if seek offsets are in the range of 0 to LLONG_MAX. + * + * Parameters: fl Flock structure. + * cur_file_offset Current offset in the file. + * + * Returns: 0 on Success. + * EOVERFLOW on overflow. + * EINVAL on offset less than zero. + */ + +static int +check_file_seek_range(struct flock *fl, off_t cur_file_offset) +{ + if (fl->l_whence == SEEK_CUR) { + /* Check if the start marker is beyond LLONG_MAX. */ + if (CHECK_ADD_OVERFLOW_INT64L(fl->l_start, cur_file_offset)) { + /* Check if start marker is negative */ + if (fl->l_start < 0) { + return EINVAL; + } + return EOVERFLOW; + } + /* Check if the start marker is negative. */ + if (fl->l_start + cur_file_offset < 0) { + return EINVAL; + } + /* 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))) { + return EOVERFLOW; + } + /* Check if the end marker is negative. */ + if ((fl->l_len <= 0) && (fl->l_start + cur_file_offset + + fl->l_len < 0)) { + return EINVAL; + } + } else if (fl->l_whence == SEEK_SET) { + /* Check if the start marker is negative. */ + if (fl->l_start < 0) { + return EINVAL; + } + /* Check if the end marker is beyond LLONG_MAX. */ + if ((fl->l_len > 0) && + CHECK_ADD_OVERFLOW_INT64L(fl->l_start, fl->l_len - 1)) { + return EOVERFLOW; + } + /* Check if the end marker is negative. */ + if ((fl->l_len < 0) && fl->l_start + fl->l_len < 0) { + return EINVAL; + } + } + return 0; +} + + +/* + * 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 file_lock_init(void) { - /* allocate file lock group attribute and group */ file_lck_grp_attr= lck_grp_attr_alloc_init(); - lck_grp_attr_setstat(file_lck_grp_attr); file_lck_grp = lck_grp_alloc_init("file", file_lck_grp_attr); /* Allocate file lock attribute */ file_lck_attr = lck_attr_alloc_init(); - //lck_attr_setdebug(file_lck_attr); uipc_lock = lck_mtx_alloc_init(file_lck_grp, file_lck_attr); - file_iterate_lcok = lck_mtx_alloc_init(file_lck_grp, file_lck_attr); - file_flist_lock = lck_mtx_alloc_init(file_lck_grp, file_lck_attr); +} - +/* + * proc_fdlock, proc_fdlock_spin + * + * Description: Lock to control access to the per process struct fileproc + * and struct filedesc + * + * Parameters: p Process to take the lock on + * + * Returns: void + * + * Notes: The lock is initialized in forkproc() and destroyed in + * reap_child_process(). + */ +void +proc_fdlock(proc_t p) +{ + lck_mtx_lock(&p->p_fdmlock); } +void +proc_fdlock_spin(proc_t p) +{ + lck_mtx_lock_spin(&p->p_fdmlock); +} void -proc_fdlock(struct proc *p) +proc_fdlock_assert(proc_t p, int assertflags) { - lck_mtx_lock(&p->p_fdmlock); + lck_mtx_assert(&p->p_fdmlock, assertflags); } + +/* + * proc_fdunlock + * + * Description: Unlock the lock previously locked by a call to proc_fdlock() + * + * Parameters: p Process to drop the lock on + * + * Returns: void + */ void -proc_fdunlock(struct proc *p) +proc_fdunlock(proc_t p) { lck_mtx_unlock(&p->p_fdmlock); } + /* * System calls on descriptors. */ + +/* + * getdtablesize + * + * Description: Returns the per process maximum size of the descriptor table + * + * Parameters: p Process being queried + * retval Pointer to the call return area + * + * Returns: 0 Success + * + * Implicit returns: + * *retval (modified) Size of dtable + */ int -getdtablesize(struct proc *p, __unused struct getdtablesize_args *uap, register_t *retval) +getdtablesize(proc_t p, __unused struct getdtablesize_args *uap, int32_t *retval) { - proc_fdlock(p); + proc_fdlock_spin(p); *retval = min((int)p->p_rlimit[RLIMIT_NOFILE].rlim_cur, maxfiles); proc_fdunlock(p); return (0); } -int -ogetdtablesize(struct proc *p, __unused void *uap, register_t *retval) + +void +procfdtbl_reservefd(struct proc * p, int fd) { - proc_fdlock(p); - *retval = min((int)p->p_rlimit[RLIMIT_NOFILE].rlim_cur, NOFILE); - proc_fdunlock(p); + p->p_fd->fd_ofiles[fd] = NULL; + p->p_fd->fd_ofileflags[fd] |= UF_RESERVED; +} - return (0); +void +procfdtbl_markclosefd(struct proc * p, int fd) +{ + p->p_fd->fd_ofileflags[fd] |= (UF_RESERVED | UF_CLOSING); +} + +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 ((p->p_fd->fd_ofileflags[fd] & UF_RESVWAIT) == UF_RESVWAIT) { + p->p_fd->fd_ofileflags[fd] &= ~UF_RESVWAIT; + wakeup(&p->p_fd); + } +} + +void +procfdtbl_waitfd(struct proc * p, int fd) +{ + p->p_fd->fd_ofileflags[fd] |= UF_RESVWAIT; + msleep(&p->p_fd, &p->p_fdmlock, PRIBIO, "ftbl_waitfd", NULL); +} + + +void +procfdtbl_clearfd(struct proc * p, int fd) +{ + int waiting; + + 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) { + wakeup(&p->p_fd); + } } -static __inline__ void -_fdrelse(struct filedesc *fdp, int fd) +/* + * _fdrelse + * + * Description: Inline utility function to free an fd in a filedesc + * + * Parameters: fdp Pointer to filedesc fd lies in + * fd fd to free + * reserv fd should be reserved + * + * Returns: void + * + * Locks: Assumes proc_fdlock for process pointing to fdp is held by + * the caller + */ +static void +_fdrelse(struct proc * p, int fd) { + struct filedesc *fdp = p->p_fd; + int nfd = 0; + if (fd < fdp->fd_freefile) fdp->fd_freefile = fd; #if DIAGNOSTIC if (fd > fdp->fd_lastfile) - panic("fdrelse: fd_lastfile inconsistent"); + panic("fdrelse: fd_lastfile inconsistent"); #endif - fdp->fd_ofiles[fd] = NULL; - fdp->fd_ofileflags[fd] = 0; + procfdtbl_clearfd(p, fd); - while ((fd = fdp->fd_lastfile) > 0 && - fdp->fd_ofiles[fd] == NULL && - !(fdp->fd_ofileflags[fd] & UF_RESERVED)) + while ((nfd = fdp->fd_lastfile) > 0 && + fdp->fd_ofiles[nfd] == NULL && + !(fdp->fd_ofileflags[nfd] & UF_RESERVED)) fdp->fd_lastfile--; } + +int +fd_rdwr( + int fd, + enum uio_rw rw, + uint64_t base, + int64_t len, + enum uio_seg segflg, + off_t offset, + int io_flg, + int64_t *aresid) +{ + struct fileproc *fp; + proc_t p; + int error = 0; + int flags = 0; + int spacetype; + uio_t auio = NULL; + char uio_buf[ UIO_SIZEOF(1) ]; + struct vfs_context context = *(vfs_context_current()); + bool wrote_some = false; + + p = current_proc(); + + 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) { + error = EINVAL; + goto out; + } + if (rw == UIO_WRITE && !(fp->f_flag & FWRITE)) { + error = EBADF; + goto out; + } + + if (rw == UIO_READ && !(fp->f_flag & FREAD)) { + error = EBADF; + goto out; + } + + context.vc_ucred = fp->f_fglob->fg_cred; + + if (UIO_SEG_IS_USER_SPACE(segflg)) + spacetype = proc_is64bit(p) ? UIO_USERSPACE64 : UIO_USERSPACE32; + else + spacetype = UIO_SYSSPACE; + + auio = uio_createwithbuffer(1, offset, spacetype, rw, &uio_buf[0], sizeof(uio_buf)); + + uio_addiov(auio, base, len); + + if ( !(io_flg & IO_APPEND)) + flags = FOF_OFFSET; + + 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 + error = fo_read(fp, auio, flags, &context); + + if (aresid) + *aresid = uio_resid(auio); + else { + if (uio_resid(auio) && error == 0) + error = EIO; + } +out: + if (wrote_some) + fp_drop_written(p, fd, fp); + else + fp_drop(p, fd, fp, 0); + + return error; +} + + + /* - * Duplicate a file descriptor. + * dup + * + * Description: Duplicate a file descriptor. + * + * Parameters: p Process performing the dup + * uap->fd The fd to dup + * retval Pointer to the call return area + * + * Returns: 0 Success + * !0 Errno + * + * Implicit returns: + * *retval (modified) The new descriptor */ -/* ARGSUSED */ int -dup(p, uap, retval) - struct proc *p; - struct dup_args *uap; - register_t *retval; +dup(proc_t p, struct dup_args *uap, int32_t *retval) { - register struct filedesc *fdp = p->p_fd; - register int old = uap->fd; + struct filedesc *fdp = p->p_fd; + int old = uap->fd; int new, error; struct fileproc *fp; @@ -246,41 +530,68 @@ dup(p, uap, retval) proc_fdunlock(p); return(error); } + 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); + return (error); + } if ( (error = fdalloc(p, 0, &new)) ) { fp_drop(p, old, fp, 1); proc_fdunlock(p); return (error); } - error = finishdup(p, fdp, old, new, retval); + 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) { + KERNEL_ENERGYTRACE(kEnTrActKernSocket, DBG_FUNC_START, + new, 0, (int64_t)VM_KERNEL_ADDRPERM(fp->f_data)); + } + return (error); } /* - * Duplicate a file descriptor to a particular value. + * dup2 + * + * Description: Duplicate a file descriptor to a particular value. + * + * Parameters: p Process performing the dup + * uap->from The fd to dup + * uap->to The fd to dup it to + * retval Pointer to the call return area + * + * Returns: 0 Success + * !0 Errno + * + * Implicit returns: + * *retval (modified) The new descriptor */ -/* ARGSUSED */ int -dup2(p, uap, retval) - struct proc *p; - struct dup2_args *uap; - register_t *retval; +dup2(proc_t p, struct dup2_args *uap, int32_t *retval) { - register struct filedesc *fdp = p->p_fd; - register int old = uap->from, new = uap->to; + struct filedesc *fdp = p->p_fd; + int old = uap->from, new = uap->to; int i, error; - struct fileproc *fp; + struct fileproc *fp, *nfp; proc_fdlock(p); +startover: if ( (error = fp_lookup(p, old, &fp, 1)) ) { proc_fdunlock(p); return(error); } + 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); + return (error); + } if (new < 0 || - new >= p->p_rlimit[RLIMIT_NOFILE].rlim_cur || + (rlim_t)new >= p->p_rlimit[RLIMIT_NOFILE].rlim_cur || new >= maxfiles) { fp_drop(p, old, fp, 1); proc_fdunlock(p); @@ -299,60 +610,165 @@ dup2(p, uap, retval) return (error); } if (new != i) { - _fdrelse(fdp, i); + fdrelse(p, i); goto closeit; } } else { - struct fileproc **fpp; - char flags; closeit: - flags = fdp->fd_ofileflags[new]; - if ((flags & (UF_RESERVED | UF_CLOSING)) == UF_RESERVED) { + while ((fdp->fd_ofileflags[new] & UF_RESERVED) == UF_RESERVED) { + fp_drop(p, old, fp, 1); + procfdtbl_waitfd(p, new); +#if DIAGNOSTIC + proc_fdlock_assert(p, LCK_MTX_ASSERT_OWNED); +#endif + goto startover; + } + + if ((fdp->fd_ofiles[new] != NULL) && + ((error = fp_lookup(p, new, &nfp, 1)) == 0)) { fp_drop(p, old, fp, 1); - proc_fdunlock(p); - return (EBADF); + if (FP_ISGUARDED(nfp, GUARD_CLOSE)) { + error = fp_guard_exception(p, + new, nfp, kGUARD_EXC_CLOSE); + (void) fp_drop(p, new, nfp, 1); + proc_fdunlock(p); + 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 { +#if DIAGNOSTIC + if (fdp->fd_ofiles[new] != NULL) + panic("dup2: no ref on fileproc %d", new); +#endif + procfdtbl_reservefd(p, new); } - /* - * dup2() must succeed even if the close has an error. - */ - if (*(fpp = &fdp->fd_ofiles[new])) { - struct fileproc *nfp = *fpp; +#if DIAGNOSTIC + proc_fdlock_assert(p, LCK_MTX_ASSERT_OWNED); +#endif - close_internal(p, new, nfp, (CLOSEINT_LOCKED | CLOSEINT_WAITONCLOSE | CLOSEINT_NOFDRELSE | CLOSEINT_NOFDNOREF)); - *fpp = NULL; - } } - error = finishdup(p, fdp, old, new, retval); +#if DIAGNOSTIC + if (fdp->fd_ofiles[new] != 0) + panic("dup2: overwriting fd_ofiles with new %d", new); + if ((fdp->fd_ofileflags[new] & UF_RESERVED) == 0) + 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); return(error); } + +/* + * fcntl + * + * Description: The file control system call. + * + * Parameters: p Process performing the fcntl + * uap->fd The fd to operate against + * uap->cmd The command to perform + * uap->arg Pointer to the command argument + * retval Pointer to the call return area + * + * Returns: 0 Success + * !0 Errno (see fcntl_nocancel) + * + * Implicit returns: + * *retval (modified) fcntl return value (if any) + * + * Notes: This system call differs from fcntl_nocancel() in that it + * tests for cancellation prior to performing a potentially + * blocking operation. + */ +int +fcntl(proc_t p, struct fcntl_args *uap, int32_t *retval) +{ + __pthread_testcancel(1); + return(fcntl_nocancel(p, (struct fcntl_nocancel_args *)uap, retval)); +} + + /* - * The file control system call. + * fcntl_nocancel + * + * Description: A non-cancel-testing file control system call. + * + * Parameters: p Process performing the fcntl + * uap->fd The fd to operate against + * uap->cmd The command to perform + * uap->arg Pointer to the command argument + * retval Pointer to the call return area + * + * Returns: 0 Success + * EINVAL + * fp_lookup:EBADF Bad file descriptor + * [F_DUPFD] + * fdalloc:EMFILE + * fdalloc:ENOMEM + * finishdup:EBADF + * finishdup:ENOMEM + * [F_SETOWN] + * ESRCH + * [F_SETLK] + * EBADF + * EOVERFLOW + * copyin:EFAULT + * vnode_getwithref:??? + * VNOP_ADVLOCK:??? + * msleep:ETIMEDOUT + * [F_GETLK] + * EBADF + * EOVERFLOW + * copyin:EFAULT + * copyout:EFAULT + * vnode_getwithref:??? + * VNOP_ADVLOCK:??? + * [F_PREALLOCATE] + * EBADF + * EINVAL + * copyin:EFAULT + * copyout:EFAULT + * vnode_getwithref:??? + * VNOP_ALLOCATE:??? + * [F_SETSIZE,F_RDADVISE] + * EBADF + * copyin:EFAULT + * vnode_getwithref:??? + * [F_RDAHEAD,F_NOCACHE] + * EBADF + * vnode_getwithref:??? + * [???] + * + * Implicit returns: + * *retval (modified) fcntl return value (if any) */ int -fcntl(p, uap, retval) - struct proc *p; - struct fcntl_args *uap; - register_t *retval; +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; - struct vnode *vp; - int i, tmp, error, error2, flg = F_POSIX; + struct vnode *vp = NULLVP; /* for AUDIT_ARG() at end */ + int i, tmp, error, error2, flg = 0; struct flock fl; + struct flocktimeout fltimeout; + struct timespec *timeout = NULL; struct vfs_context context; off_t offset; int newmin; daddr64_t lbn, bn; - int devBlockSize = 0; unsigned int fflag; user_addr_t argp; + boolean_t is64bit; AUDIT_ARG(fd, uap->fd); AUDIT_ARG(cmd, uap->cmd); @@ -362,28 +778,45 @@ fcntl(p, uap, retval) proc_fdunlock(p); return(error); } - context.vc_proc = p; + context.vc_thread = current_thread(); context.vc_ucred = fp->f_cred; - if (proc_is64bit(p)) { + + is64bit = proc_is64bit(p); + if (is64bit) { argp = uap->arg; } 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 sign extension - * issues. Our sys call munger will sign extend a long when we are - * called from a 32-bit process. Since we can never have an address - * greater than 32-bits from a 32-bit process we lop off the top - * 32-bits to avoid getting the wrong address + /* + * Since the arg parameter is defined as a long but may be + * either a long or a pointer we must take care to handle + * sign extension issues. Our sys call munger will sign + * extend a long when we are called from a 32-bit process. + * Since we can never have an address greater than 32-bits + * from a 32-bit process we lop off the top 32-bits to avoid + * getting the wrong address */ - argp = CAST_USER_ADDR_T(uap->arg); + argp = CAST_USER_ADDR_T((uint32_t)uap->arg); } pop = &fdp->fd_ofileflags[fd]; +#if CONFIG_MACF + error = mac_file_check_fcntl(proc_ucred(p), fp->f_fglob, uap->cmd, + uap->arg); + if (error) + goto out; +#endif + switch (uap->cmd) { case F_DUPFD: - newmin = CAST_DOWN(int, uap->arg); + case F_DUPFD_CLOEXEC: + 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); if ((u_int)newmin >= p->p_rlimit[RLIMIT_NOFILE].rlim_cur || newmin >= maxfiles) { error = EINVAL; @@ -391,17 +824,27 @@ fcntl(p, uap, retval) } if ( (error = fdalloc(p, newmin, &i)) ) goto out; - error = finishdup(p, fdp, fd, i, retval); + error = finishdup(p, fdp, fd, i, + uap->cmd == F_DUPFD_CLOEXEC ? UF_EXCLOSE : 0, retval); goto out; case F_GETFD: - *retval = (*pop & UF_EXCLOSE)? 1 : 0; + *retval = (*pop & UF_EXCLOSE)? FD_CLOEXEC : 0; error = 0; goto out; case F_SETFD: - *pop = (*pop &~ UF_EXCLOSE) | - (uap->arg & 1)? UF_EXCLOSE : 0; + AUDIT_ARG(value32, uap->arg); + if (uap->arg & FD_CLOEXEC) + *pop |= UF_EXCLOSE; + else { + if (FILEPROC_TYPE(fp) == FTYPE_GUARDED) { + error = fp_guard_exception(p, + fd, fp, kGUARD_EXC_NOCLOEXEC); + goto out; + } + *pop &= ~UF_EXCLOSE; + } error = 0; goto out; @@ -412,19 +855,20 @@ fcntl(p, uap, retval) case F_SETFL: fp->f_flag &= ~FCNTLFLAGS; - tmp = CAST_DOWN(int, uap->arg); + 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; - error = fo_ioctl(fp, FIONBIO, (caddr_t)&tmp, p); + error = fo_ioctl(fp, FIONBIO, (caddr_t)&tmp, &context); if (error) goto out; tmp = fp->f_flag & FASYNC; - error = fo_ioctl(fp, FIOASYNC, (caddr_t)&tmp, p); + error = fo_ioctl(fp, FIOASYNC, (caddr_t)&tmp, &context); if (!error) goto out; fp->f_flag &= ~FNONBLOCK; tmp = 0; - (void)fo_ioctl(fp, FIONBIO, (caddr_t)&tmp, p); + (void)fo_ioctl(fp, FIONBIO, (caddr_t)&tmp, &context); goto out; case F_GETOWN: @@ -433,40 +877,120 @@ fcntl(p, uap, retval) error = 0; goto out; } - error = fo_ioctl(fp, (int)TIOCGPGRP, (caddr_t)retval, p); + error = fo_ioctl(fp, (int)TIOCGPGRP, (caddr_t)retval, &context); *retval = -*retval; goto out; case F_SETOWN: - tmp = CAST_DOWN(pid_t, uap->arg); + tmp = CAST_DOWN_EXPLICIT(pid_t, uap->arg); /* arg is an int, so we won't lose bits */ + AUDIT_ARG(value32, tmp); if (fp->f_type == DTYPE_SOCKET) { ((struct socket *)fp->f_data)->so_pgid = tmp; error =0; goto out; } if (fp->f_type == DTYPE_PIPE) { - error = fo_ioctl(fp, (int)TIOCSPGRP, (caddr_t)&tmp, p); + error = fo_ioctl(fp, TIOCSPGRP, (caddr_t)&tmp, &context); goto out; } if (tmp <= 0) { tmp = -tmp; } else { - struct proc *p1 = pfind(tmp); + proc_t p1 = proc_find(tmp); if (p1 == 0) { error = ESRCH; goto out; } - tmp = (int)p1->p_pgrp->pg_id; + tmp = (int)p1->p_pgrpid; + proc_rele(p1); + } + error = fo_ioctl(fp, (int)TIOCSPGRP, (caddr_t)&tmp, &context); + goto out; + + case F_SETNOSIGPIPE: + tmp = CAST_DOWN_EXPLICIT(int, uap->arg); + if (fp->f_type == DTYPE_SOCKET) { +#if SOCKETS + error = sock_setsockopt((struct socket *)fp->f_data, + SOL_SOCKET, SO_NOSIGPIPE, &tmp, sizeof (tmp)); +#else + error = EINVAL; +#endif + } else { + struct fileglob *fg = fp->f_fglob; + + lck_mtx_lock_spin(&fg->fg_lock); + if (tmp) + fg->fg_lflags |= FG_NOSIGPIPE; + else + fg->fg_lflags &= FG_NOSIGPIPE; + lck_mtx_unlock(&fg->fg_lock); + error = 0; + } + goto out; + + case F_GETNOSIGPIPE: + if (fp->f_type == DTYPE_SOCKET) { +#if SOCKETS + int retsize = sizeof (*retval); + 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; + error = 0; + } + goto out; + + case F_SETCONFINED: + /* + * If this is the only reference to this fglob in the process + * and it's already marked as close-on-fork then mark it as + * (immutably) "confined" i.e. any fd that points to it will + * forever be close-on-fork, and attempts to use an IPC + * mechanism to move the descriptor elsewhere will fail. + */ + if (CAST_DOWN_EXPLICIT(int, uap->arg)) { + struct fileglob *fg = fp->f_fglob; + + lck_mtx_lock_spin(&fg->fg_lock); + if (fg->fg_lflags & FG_CONFINED) + error = 0; + else if (1 != fg->fg_count) + error = EAGAIN; /* go close the dup .. */ + else if (UF_FORKCLOSE == (*pop & UF_FORKCLOSE)) { + fg->fg_lflags |= FG_CONFINED; + error = 0; + } 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. + */ + error = EPERM; /* immutable */ } - error = fo_ioctl(fp, (int)TIOCSPGRP, (caddr_t)&tmp, p); goto out; + case F_GETCONFINED: + *retval = (fp->f_fglob->fg_lflags & FG_CONFINED) ? 1 : 0; + error = 0; + goto out; + + case F_SETLKWTIMEOUT: case F_SETLKW: + case F_OFD_SETLKWTIMEOUT: + case F_OFD_SETLKW: flg |= F_WAIT; /* Fall into F_SETLK */ case F_SETLK: + case F_OFD_SETLK: if (fp->f_type != DTYPE_VNODE) { error = EBADF; goto out; @@ -478,53 +1002,126 @@ fcntl(p, uap, retval) proc_fdunlock(p); /* Copy in the lock structure */ - error = copyin(argp, (caddr_t)&fl, sizeof (fl)); + if (F_SETLKWTIMEOUT == uap->cmd || + F_OFD_SETLKWTIMEOUT == uap->cmd) { + error = copyin(argp, (caddr_t) &fltimeout, sizeof(fltimeout)); + if (error) { + goto outdrop; + } + fl = fltimeout.fl; + timeout = &fltimeout.timeout; + } else { + error = copyin(argp, (caddr_t)&fl, sizeof(fl)); + if (error) { + goto outdrop; + } + } + + /* Check starting byte and ending byte for EOVERFLOW in SEEK_CUR */ + /* and ending byte for EOVERFLOW in SEEK_SET */ + error = check_file_seek_range(&fl, offset); if (error) { goto outdrop; } + if ( (error = vnode_getwithref(vp)) ) { goto outdrop; } if (fl.l_whence == SEEK_CUR) fl.l_start += offset; - switch (fl.l_type) { - - case F_RDLCK: - if ((fflag & FREAD) == 0) { - (void)vnode_put(vp); - error = EBADF; - goto outdrop; - } - OSBitOrAtomic(P_LADVLOCK, &p->p_ladvflag); - error = VNOP_ADVLOCK(vp, (caddr_t)p, F_SETLK, &fl, flg, &context); - (void)vnode_put(vp); - goto outdrop; - - case F_WRLCK: - if ((fflag & FWRITE) == 0) { - (void)vnode_put(vp); - error = EBADF; - goto outdrop; - } - OSBitOrAtomic(P_LADVLOCK, &p->p_ladvflag); - error = VNOP_ADVLOCK(vp, (caddr_t)p, F_SETLK, &fl, flg, &context); - (void)vnode_put(vp); - goto outdrop; - - case F_UNLCK: - error = VNOP_ADVLOCK(vp, (caddr_t)p, F_UNLCK, &fl, - F_POSIX, &context); - (void)vnode_put(vp); - goto outdrop; - - default: +#if CONFIG_MACF + error = mac_file_check_lock(proc_ucred(p), fp->f_fglob, + F_SETLK, &fl); + if (error) { (void)vnode_put(vp); - error = EINVAL; goto outdrop; } - +#endif + switch (uap->cmd) { + case F_OFD_SETLK: + case F_OFD_SETLKW: + case F_OFD_SETLKWTIMEOUT: + flg |= F_OFD_LOCK; + switch (fl.l_type) { + case F_RDLCK: + if ((fflag & FREAD) == 0) { + error = EBADF; + break; + } + error = VNOP_ADVLOCK(vp, (caddr_t)fp->f_fglob, + F_SETLK, &fl, flg, &context, timeout); + break; + case F_WRLCK: + if ((fflag & FWRITE) == 0) { + error = EBADF; + break; + } + error = VNOP_ADVLOCK(vp, (caddr_t)fp->f_fglob, + F_SETLK, &fl, flg, &context, timeout); + break; + case F_UNLCK: + error = VNOP_ADVLOCK(vp, (caddr_t)fp->f_fglob, + F_UNLCK, &fl, F_OFD_LOCK, &context, + timeout); + break; + default: + error = EINVAL; + break; + } + if (0 == error && + (F_RDLCK == fl.l_type || F_WRLCK == fl.l_type)) { + struct fileglob *fg = fp->f_fglob; + + /* + * arrange F_UNLCK on last close (once + * set, FG_HAS_OFDLOCK is immutable) + */ + if ((fg->fg_lflags & FG_HAS_OFDLOCK) == 0) { + lck_mtx_lock_spin(&fg->fg_lock); + fg->fg_lflags |= FG_HAS_OFDLOCK; + lck_mtx_unlock(&fg->fg_lock); + } + } + break; + default: + flg |= F_POSIX; + switch (fl.l_type) { + case F_RDLCK: + if ((fflag & FREAD) == 0) { + error = EBADF; + break; + } + // XXX UInt32 unsafe for LP64 kernel + OSBitOrAtomic(P_LADVLOCK, &p->p_ladvflag); + error = VNOP_ADVLOCK(vp, (caddr_t)p, + F_SETLK, &fl, flg, &context, timeout); + break; + case F_WRLCK: + if ((fflag & FWRITE) == 0) { + error = EBADF; + break; + } + // XXX UInt32 unsafe for LP64 kernel + OSBitOrAtomic(P_LADVLOCK, &p->p_ladvflag); + error = VNOP_ADVLOCK(vp, (caddr_t)p, + F_SETLK, &fl, flg, &context, timeout); + break; + case F_UNLCK: + error = VNOP_ADVLOCK(vp, (caddr_t)p, + F_UNLCK, &fl, F_POSIX, &context, timeout); + break; + default: + error = EINVAL; + break; + } + break; + } + (void) vnode_put(vp); + goto outdrop; + case F_GETLK: + case F_OFD_GETLK: if (fp->f_type != DTYPE_VNODE) { error = EBADF; goto out; @@ -535,20 +1132,70 @@ fcntl(p, uap, retval) proc_fdunlock(p); /* Copy in the lock structure */ - error = copyin(argp, (caddr_t)&fl, sizeof (fl)); + error = copyin(argp, (caddr_t)&fl, sizeof(fl)); if (error) goto outdrop; + /* Check starting byte and ending byte for EOVERFLOW in SEEK_CUR */ + /* and ending byte for EOVERFLOW in SEEK_SET */ + error = check_file_seek_range(&fl, offset); + if (error) { + goto outdrop; + } + + if ((fl.l_whence == SEEK_SET) && (fl.l_start < 0)) { + error = EINVAL; + goto outdrop; + } + + switch (fl.l_type) { + case F_RDLCK: + case F_UNLCK: + case F_WRLCK: + break; + default: + error = EINVAL; + goto outdrop; + } + + switch (fl.l_whence) { + case SEEK_CUR: + case SEEK_SET: + case SEEK_END: + break; + default: + error = EINVAL; + goto outdrop; + } + if ( (error = vnode_getwithref(vp)) == 0 ) { if (fl.l_whence == SEEK_CUR) fl.l_start += offset; - error = VNOP_ADVLOCK(vp, (caddr_t)p, F_GETLK, &fl, F_POSIX, &context); +#if CONFIG_MACF + error = mac_file_check_lock(proc_ucred(p), fp->f_fglob, + uap->cmd, &fl); + if (error == 0) +#endif + switch (uap->cmd) { + case F_OFD_GETLK: + error = VNOP_ADVLOCK(vp, (caddr_t)fp->f_fglob, + F_GETLK, &fl, F_OFD_LOCK, &context, NULL); + break; + case F_OFD_GETLKPID: + error = VNOP_ADVLOCK(vp, (caddr_t)fp->f_fglob, + F_GETLKPID, &fl, F_OFD_LOCK, &context, NULL); + break; + default: + error = VNOP_ADVLOCK(vp, (caddr_t)p, + uap->cmd, &fl, F_POSIX, &context, NULL); + break; + } (void)vnode_put(vp); if (error == 0) - error = copyout((caddr_t)&fl, argp, sizeof (fl)); + error = copyout((caddr_t)&fl, argp, sizeof(fl)); } goto outdrop; @@ -570,7 +1217,7 @@ fcntl(p, uap, retval) goto outdrop; } - error = copyin(argp, (caddr_t)&alloc_struct, sizeof (alloc_struct)); + error = copyin(argp, (caddr_t)&alloc_struct, sizeof(alloc_struct)); if (error) goto outdrop; @@ -630,7 +1277,7 @@ fcntl(p, uap, retval) &context); (void)vnode_put(vp); - error2 = copyout((caddr_t)&alloc_struct, argp, sizeof (alloc_struct)); + error2 = copyout((caddr_t)&alloc_struct, argp, sizeof(alloc_struct)); if (error == 0) error = error2; @@ -643,32 +1290,42 @@ fcntl(p, uap, retval) error = EBADF; goto out; } + vp = (struct vnode *)fp->f_data; proc_fdunlock(p); error = copyin(argp, (caddr_t)&offset, sizeof (off_t)); if (error) goto outdrop; + AUDIT_ARG(value64, offset); + + error = vnode_getwithref(vp); + if (error) + goto outdrop; +#if CONFIG_MACF + error = mac_vnode_check_truncate(&context, + fp->f_fglob->fg_cred, vp); + if (error) { + (void)vnode_put(vp); + goto outdrop; + } +#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 */ - - if (!is_suser()) { + if (!kauth_cred_issuser(kauth_cred_get())) { error = EACCES; - goto outdrop; - } - vp = (struct vnode *)fp->f_data; - - if ( (error = vnode_getwithref(vp)) == 0 ) { - /* + } else { + /* * set the file size */ - error = vnode_setsize(vp, offset, IO_NOZEROFILL, &context); - - (void)vnode_put(vp); + error = vnode_setsize(vp, offset, IO_NOZEROFILL, + &context); } + + (void)vnode_put(vp); goto outdrop; case F_RDAHEAD: @@ -676,29 +1333,62 @@ fcntl(p, uap, retval) error = EBADF; goto out; } - vp = (struct vnode *)fp->f_data; - proc_fdunlock(p); + if (uap->arg) + fp->f_fglob->fg_flag &= ~FNORDAHEAD; + else + fp->f_fglob->fg_flag |= FNORDAHEAD; - if ( (error = vnode_getwithref(vp)) == 0) { - if (uap->arg) - vnode_clearnoreadahead(vp); - else - vnode_setnoreadahead(vp); + goto out; - (void)vnode_put(vp); + case F_NOCACHE: + if (fp->f_type != DTYPE_VNODE) { + error = EBADF; + goto out; } - goto outdrop; + if (uap->arg) + fp->f_fglob->fg_flag |= FNOCACHE; + else + fp->f_fglob->fg_flag &= ~FNOCACHE; - case F_NOCACHE: + goto out; + + case F_NODIRECT: + if (fp->f_type != DTYPE_VNODE) { + error = EBADF; + goto out; + } + if (uap->arg) + fp->f_fglob->fg_flag |= FNODIRECT; + else + fp->f_fglob->fg_flag &= ~FNODIRECT; + + goto out; + + case F_SINGLE_WRITER: if (fp->f_type != DTYPE_VNODE) { error = EBADF; goto out; } + if (uap->arg) + fp->f_fglob->fg_flag |= FSINGLE_WRITER; + else + fp->f_fglob->fg_flag &= ~FSINGLE_WRITER; + + goto out; + + case F_GLOBAL_NOCACHE: + 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 ) { - if (uap->arg) + + *retval = vnode_isnocache(vp); + + if (uap->arg) vnode_setnocache(vp); else vnode_clearnocache(vp); @@ -707,9 +1397,9 @@ fcntl(p, uap, retval) } goto outdrop; - case F_GLOBAL_NOCACHE: - if (fp->f_type != DTYPE_VNODE) { - error = EBADF; + case F_CHECK_OPENEVT: + if (fp->f_type != DTYPE_VNODE) { + error = EBADF; goto out; } vp = (struct vnode *)fp->f_data; @@ -717,12 +1407,12 @@ fcntl(p, uap, retval) if ( (error = vnode_getwithref(vp)) == 0 ) { - *retval = vnode_isnocache(vp); + *retval = vnode_is_openevt(vp); - if (uap->arg) - vnode_setnocache(vp); + if (uap->arg) + vnode_set_openevt(vp); else - vnode_clearnocache(vp); + vnode_clear_openevt(vp); (void)vnode_put(vp); } @@ -738,7 +1428,7 @@ fcntl(p, uap, retval) 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; if ( (error = vnode_getwithref(vp)) == 0 ) { error = VNOP_IOCTL(vp, F_RDADVISE, (caddr_t)&ra_struct, 0, &context); @@ -748,59 +1438,39 @@ fcntl(p, uap, retval) goto outdrop; } - case F_READBOOTSTRAP: - case F_WRITEBOOTSTRAP: { - fbootstraptransfer_t fbt_struct; - user_fbootstraptransfer_t user_fbt_struct; - int sizeof_struct; - caddr_t boot_structp; + 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 (IS_64BIT_PROCESS(p)) { - sizeof_struct = sizeof(user_fbt_struct); - boot_structp = (caddr_t) &user_fbt_struct; - } - else { - sizeof_struct = sizeof(fbt_struct); - boot_structp = (caddr_t) &fbt_struct; - } - error = copyin(argp, boot_structp, sizeof_struct); - if (error) - goto outdrop; - if ( (error = vnode_getwithref(vp)) ) { - goto outdrop; - } - if (uap->cmd == F_WRITEBOOTSTRAP) { - /* - * Make sure that we are root. Updating the - * bootstrap on a disk could be a security hole - */ - if (!is_suser()) { - (void)vnode_put(vp); - error = EACCES; - goto outdrop; - } - } - if (strcmp(vnode_mount(vp)->mnt_vfsstat.f_fstypename, "hfs") != 0) { - error = EINVAL; - } else { - /* - * call vnop_ioctl to handle the I/O - */ - error = VNOP_IOCTL(vp, uap->cmd, boot_structp, 0, &context); - } - (void)vnode_put(vp); - goto outdrop; - } - case F_LOG2PHYS: { + if ( (error = vnode_getwithref(vp)) == 0 ) { + error = cluster_push(vp, 0); + + (void)vnode_put(vp); + } + goto outdrop; + + case F_LOG2PHYS: + case F_LOG2PHYS_EXT: { struct log2phys l2p_struct; /* structure for allocate command */ + int devBlockSize; + + off_t file_offset = 0; + size_t a_size = 0; + size_t run = 0; + if (uap->cmd == F_LOG2PHYS_EXT) { + error = copyin(argp, (caddr_t)&l2p_struct, sizeof(l2p_struct)); + if (error) + goto out; + file_offset = l2p_struct.l2p_devoffset; + } else { + file_offset = fp->f_offset; + } if (fp->f_type != DTYPE_VNODE) { error = EBADF; goto out; @@ -810,7 +1480,7 @@ fcntl(p, uap, retval) if ( (error = vnode_getwithref(vp)) ) { goto outdrop; } - error = VNOP_OFFTOBLK(vp, fp->f_offset, &lbn); + error = VNOP_OFFTOBLK(vp, file_offset, &lbn); if (error) { (void)vnode_put(vp); goto outdrop; @@ -821,17 +1491,43 @@ fcntl(p, uap, retval) goto outdrop; } devBlockSize = vfs_devblocksize(vnode_mount(vp)); + if (uap->cmd == F_LOG2PHYS_EXT) { + if (l2p_struct.l2p_contigbytes < 0) { + vnode_put(vp); + error = EINVAL; + goto outdrop; + } - error = VNOP_BLOCKMAP(vp, offset, devBlockSize, &bn, NULL, NULL, 0, &context); + a_size = MIN((uint64_t)l2p_struct.l2p_contigbytes, SIZE_MAX); + } else { + a_size = devBlockSize; + } + + error = VNOP_BLOCKMAP(vp, offset, a_size, &bn, &run, NULL, 0, &context); (void)vnode_put(vp); if (!error) { l2p_struct.l2p_flags = 0; /* for now */ - l2p_struct.l2p_contigbytes = 0; /* for now */ - l2p_struct.l2p_devoffset = bn * devBlockSize; - l2p_struct.l2p_devoffset += fp->f_offset - offset; - error = copyout((caddr_t)&l2p_struct, argp, sizeof (l2p_struct)); + if (uap->cmd == F_LOG2PHYS_EXT) { + l2p_struct.l2p_contigbytes = run - (file_offset - offset); + } else { + l2p_struct.l2p_contigbytes = 0; /* for now */ + } + + /* + * The block number being -1 suggests that the file offset is not backed + * by any real blocks on-disk. As a result, just let it be passed back up wholesale. + */ + if (bn == -1) { + /* Don't multiply it by the block size */ + l2p_struct.l2p_devoffset = bn; + } + 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; } @@ -879,6 +1575,7 @@ fcntl(p, uap, retval) if ( (error = copyinstr(argp, pathbufp, MAXPATHLEN, &pathlen)) == 0 ) { if ( (error = vnode_getwithref(vp)) == 0 ) { + AUDIT_ARG(text, pathbufp); error = vn_path_package_check(vp, pathbufp, pathlen, retval); (void)vnode_put(vp); @@ -889,7 +1586,8 @@ fcntl(p, uap, retval) } case F_CHKCLEAN: // used by regression tests to see if all dirty pages got cleaned by fsync() - case F_FULLFSYNC: // fsync + flush the journal + DKIOCSYNCHRONIZECACHE + case F_FULLFSYNC: // fsync + flush the journal + DKIOCSYNCHRONIZE + case F_BARRIERFSYNC: // fsync + barrier case F_FREEZE_FS: // freeze all other fs operations for the fs of this fd case F_THAW_FS: { // thaw all frozen fs operations for the fs of this fd if (fp->f_type != DTYPE_VNODE) { @@ -906,135 +1604,1236 @@ fcntl(p, uap, retval) } break; } - - default: - if (uap->cmd < FCNTL_FS_SPECIFIC_BASE) { - error = EINVAL; - goto out; - } - // if it's a fs-specific fcntl() then just pass it through - - if (fp->f_type != DTYPE_VNODE) { + /* + * SPI (private) for opening a file starting from a dir fd + */ + case F_OPENFROM: { + struct user_fopenfrom fopen; + struct vnode_attr va; + struct nameidata nd; + int cmode; + + /* Check if this isn't a valid file descriptor */ + if ((fp->f_type != DTYPE_VNODE) || + (fp->f_flag & FREAD) == 0) { error = EBADF; goto out; } vp = (struct vnode *)fp->f_data; proc_fdunlock(p); - if ( (error = vnode_getwithref(vp)) == 0 ) { - error = VNOP_IOCTL(vp, uap->cmd, CAST_DOWN(caddr_t, argp), 0, &context); + if (vnode_getwithref(vp)) { + error = ENOENT; + goto outdrop; + } + + /* Only valid for directories */ + if (vp->v_type != VDIR) { + vnode_put(vp); + error = ENOTDIR; + goto outdrop; + } - (void)vnode_put(vp); + /* Get flags, mode and pathname arguments. */ + if (IS_64BIT_PROCESS(p)) { + error = copyin(argp, &fopen, sizeof(fopen)); + } else { + struct user32_fopenfrom fopen32; + + error = copyin(argp, &fopen32, sizeof(fopen32)); + fopen.o_flags = fopen32.o_flags; + fopen.o_mode = fopen32.o_mode; + fopen.o_pathname = CAST_USER_ADDR_T(fopen32.o_pathname); + } + if (error) { + vnode_put(vp); + goto outdrop; } + AUDIT_ARG(fflags, fopen.o_flags); + 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; + 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); + nd.ni_dvp = vp; + + error = open1(&context, &nd, fopen.o_flags, &va, + fileproc_alloc_init, NULL, retval); + + vnode_put(vp); break; - } + /* + * SPI (private) for unlinking a file starting from a dir fd + */ + case F_UNLINKFROM: { + user_addr_t pathname; -outdrop: - AUDIT_ARG(vnpath_withref, vp, ARG_VNODE1); - fp_drop(p, fd, fp, 0); - return(error); -out: - fp_drop(p, fd, fp, 1); - proc_fdunlock(p); - return(error); -} + /* Check if this isn't a valid file descriptor */ + if ((fp->f_type != DTYPE_VNODE) || + (fp->f_flag & FREAD) == 0) { + error = EBADF; + goto out; + } + vp = (struct vnode *)fp->f_data; + proc_fdunlock(p); -/* - * Common code for dup, dup2, and fcntl(F_DUPFD). - */ -int -finishdup(struct proc * p, struct filedesc *fdp, int old, int new, register_t *retval) -{ - struct fileproc *nfp; - struct fileproc *ofp; + if (vnode_getwithref(vp)) { + error = ENOENT; + goto outdrop; + } + + /* Only valid for directories */ + if (vp->v_type != VDIR) { + vnode_put(vp); + error = ENOTDIR; + goto outdrop; + } - if ((ofp = fdp->fd_ofiles[old]) == NULL || - (fdp->fd_ofileflags[old] & UF_RESERVED)) { - _fdrelse(fdp, new); - return (EBADF); - } - fg_ref(ofp); - proc_fdunlock(p); + /* Get flags, mode and pathname arguments. */ + if (IS_64BIT_PROCESS(p)) { + pathname = (user_addr_t)argp; + } else { + pathname = CAST_USER_ADDR_T(argp); + } - MALLOC_ZONE(nfp, struct fileproc *, sizeof(struct fileproc), M_FILEPROC, M_WAITOK); - bzero(nfp, sizeof(struct fileproc)); + /* Start the lookup relative to the file descriptor's vnode. */ + error = unlink1(&context, vp, pathname, UIO_USERSPACE, 0); + + vnode_put(vp); + break; - proc_fdlock(p); - nfp->f_flags = ofp->f_flags; - nfp->f_fglob = ofp->f_fglob; - nfp->f_iocount = 0; + } - fdp->fd_ofiles[new] = nfp; - fdp->fd_ofileflags[new] = fdp->fd_ofileflags[old] &~ UF_EXCLOSE; - if (new > fdp->fd_lastfile) - fdp->fd_lastfile = new; - *retval = new; - return (0); -} + case F_ADDSIGS: + case F_ADDFILESIGS: + case F_ADDFILESIGS_FOR_DYLD_SIM: + case F_ADDFILESIGS_RETURN: + { + struct cs_blob *blob = NULL; + struct user_fsignatures fs; + kern_return_t kr; + vm_offset_t kernel_blob_addr; + vm_size_t kernel_blob_size; + int blob_add_flags = 0; + if (fp->f_type != DTYPE_VNODE) { + error = EBADF; + goto out; + } + vp = (struct vnode *)fp->f_data; + proc_fdunlock(p); -int -close(struct proc *p, struct close_args *uap, __unused register_t *retval) -{ - struct fileproc *fp; - int fd = uap->fd; - int error =0; + if (uap->cmd == F_ADDFILESIGS_FOR_DYLD_SIM) { + blob_add_flags |= MAC_VNODE_CHECK_DYLD_SIM; + if ((p->p_csflags & CS_KILL) == 0) { + proc_lock(p); + p->p_csflags |= CS_KILL; + proc_unlock(p); + } + } - AUDIT_SYSCLOSE(p, fd); + error = vnode_getwithref(vp); + if (error) + goto outdrop; - proc_fdlock(p); + if (IS_64BIT_PROCESS(p)) { + error = copyin(argp, &fs, sizeof (fs)); + } else { + struct user32_fsignatures fs32; - if ( (error = fp_lookup(p,fd,&fp, 1)) ) { - proc_fdunlock(p); - return(error); - } + 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; + } - error = close_internal(p, fd, fp, CLOSEINT_LOCKED | CLOSEINT_WAITONCLOSE); + if (error) { + vnode_put(vp); + goto outdrop; + } - proc_fdunlock(p); + /* + * 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) + { + /* If this is for dyld_sim revalidate the blob */ + if (uap->cmd == F_ADDFILESIGS_FOR_DYLD_SIM) { + error = ubc_cs_blob_revalidate(vp, blob, blob_add_flags); + } - return(error); -} + } else { + /* + * An arbitrary limit, to prevent someone from mapping in a 20GB blob. This should cover + * our use cases for the immediate future, but note that at the time of this commit, some + * platforms are nearing 2MB blob sizes (with a prior soft limit of 2.5MB). + * + * We should consider how we can manage this more effectively; the above means that some + * platforms are using megabytes of memory for signing data; it merely hasn't crossed the + * threshold considered ridiculous at the time of this change. + */ +#define CS_MAX_BLOB_SIZE (40ULL * 1024ULL * 1024ULL) + if (fs.fs_blob_size > CS_MAX_BLOB_SIZE) { + error = E2BIG; + vnode_put(vp); + goto 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; + vnode_put(vp); + goto outdrop; + } + + if(uap->cmd == F_ADDSIGS) { + error = copyin(fs.fs_blob_start, + (void *) kernel_blob_addr, + kernel_blob_size); + } else /* F_ADDFILESIGS || F_ADDFILESIGS_RETURN || F_ADDFILESIGS_FOR_DYLD_SIM */ { + 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); + 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); + 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, + blob_add_flags, + &blob); + if (error) { + ubc_cs_blob_deallocate(kernel_blob_addr, + kernel_blob_size); + } else { + /* ubc_blob_add() has consumed "kernel_blob_addr" */ +#if CHECK_CS_VALIDATION_BITMAP + ubc_cs_validation_bitmap_allocate( vp ); +#endif + } + } + + if (uap->cmd == F_ADDFILESIGS_RETURN || uap->cmd == F_ADDFILESIGS_FOR_DYLD_SIM) { + /* + * 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) + end_offset = blob->csb_end_offset; + error = copyout(&end_offset, argp, sizeof (end_offset)); + } + + (void) vnode_put(vp); + break; + } + case F_FINDSIGS: { + error = ENOTSUP; + goto out; + } +#if CONFIG_PROTECT + case F_GETPROTECTIONCLASS: { + int class = 0; + + if (fp->f_type != DTYPE_VNODE) { + error = EBADF; + goto out; + } + vp = (struct vnode *)fp->f_data; + + proc_fdunlock(p); + + if (vnode_getwithref(vp)) { + error = ENOENT; + goto outdrop; + } + + error = cp_vnode_getclass (vp, &class); + if (error == 0) { + *retval = class; + } + + vnode_put(vp); + break; + } + + case F_SETPROTECTIONCLASS: { + /* tmp must be a valid PROTECTION_CLASS_* */ + tmp = CAST_DOWN_EXPLICIT(uint32_t, uap->arg); + + if (fp->f_type != DTYPE_VNODE) { + error = EBADF; + goto out; + } + vp = (struct vnode *)fp->f_data; + + proc_fdunlock(p); + + if (vnode_getwithref(vp)) { + error = ENOENT; + goto outdrop; + } + + /* 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) { + vnode_put(vp); + error = EBADF; + goto outdrop; + } + error = cp_vnode_setclass (vp, tmp); + vnode_put(vp); + break; + } + + case F_TRANSCODEKEY: { + + char *backup_keyp = NULL; + unsigned backup_key_len = CP_MAX_WRAPPEDKEYSIZE; + + if (fp->f_type != DTYPE_VNODE) { + error = EBADF; + goto out; + } + + vp = (struct vnode *)fp->f_data; + proc_fdunlock(p); + + if (vnode_getwithref(vp)) { + error = ENOENT; + goto outdrop; + } + + MALLOC(backup_keyp, char *, backup_key_len, M_TEMP, M_WAITOK); + if (backup_keyp == NULL) { + error = ENOMEM; + goto outdrop; + } + + error = cp_vnode_transcode (vp, backup_keyp, &backup_key_len); + vnode_put(vp); + + if (error == 0) { + error = copyout((caddr_t)backup_keyp, argp, backup_key_len); + *retval = backup_key_len; + } + + FREE(backup_keyp, M_TEMP); + + break; + } + + case F_GETPROTECTIONLEVEL: { + uint32_t cp_version = 0; + + if (fp->f_type != DTYPE_VNODE) { + error = EBADF; + goto out; + } + + vp = (struct vnode*) fp->f_data; + proc_fdunlock (p); + + if (vnode_getwithref(vp)) { + error = ENOENT; + goto outdrop; + } + + /* + * if cp_get_major_vers fails, error will be set to proper errno + * and cp_version will still be 0. + */ + + error = cp_get_root_major_vers (vp, &cp_version); + *retval = cp_version; + + vnode_put (vp); + break; + } + + case F_GETDEFAULTPROTLEVEL: { + uint32_t cp_default = 0; + + if (fp->f_type != DTYPE_VNODE) { + error = EBADF; + goto out; + } + + vp = (struct vnode*) fp->f_data; + proc_fdunlock (p); + + if (vnode_getwithref(vp)) { + error = ENOENT; + goto outdrop; + } + + /* + * if cp_get_major_vers fails, error will be set to proper errno + * and cp_version will still be 0. + */ + + error = cp_get_default_level(vp, &cp_default); + *retval = cp_default; + + vnode_put (vp); + break; + } + + +#endif /* CONFIG_PROTECT */ + + case F_MOVEDATAEXTENTS: { + struct fileproc *fp2 = NULL; + struct vnode *src_vp = NULLVP; + struct vnode *dst_vp = NULLVP; + /* We need to grab the 2nd FD out of the argments before moving on. */ + int fd2 = CAST_DOWN_EXPLICIT(int32_t, uap->arg); + + error = priv_check_cred(kauth_cred_get(), PRIV_VFS_MOVE_DATA_EXTENTS, 0); + if (error) + goto out; + + if (fp->f_type != DTYPE_VNODE) { + error = EBADF; + goto out; + } + + /* For now, special case HFS+ only, since this is SPI. */ + src_vp = (struct vnode *)fp->f_data; + if (src_vp->v_tag != VT_HFS) { + error = EINVAL; + goto out; + } + + /* + * 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)) ) { + error = EBADF; + goto out; + } + if (fp2->f_type != DTYPE_VNODE) { + fp_drop(p, fd2, fp2, 1); + error = EBADF; + goto out; + } + dst_vp = (struct vnode *)fp2->f_data; + if (dst_vp->v_tag != VT_HFS) { + fp_drop(p, fd2, fp2, 1); + error = EINVAL; + goto out; + } + +#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); + if (error) { + fp_drop(p, fd2, fp2, 1); + goto out; + } +#endif + /* Audit the 2nd FD */ + AUDIT_ARG(fd, fd2); + + proc_fdunlock(p); + + if (vnode_getwithref(src_vp)) { + fp_drop(p, fd2, fp2, 0); + error = ENOENT; + goto outdrop; + } + if (vnode_getwithref(dst_vp)) { + vnode_put (src_vp); + fp_drop(p, fd2, fp2, 0); + error = ENOENT; + goto outdrop; + } + + /* + * Basic asserts; validate they are not the same and that + * 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; + } + + if (dst_vp->v_mount != src_vp->v_mount) { + vnode_put (src_vp); + vnode_put (dst_vp); + fp_drop (p, fd2, fp2, 0); + error = EXDEV; + goto outdrop; + } + + /* 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) { + vnode_put(src_vp); + vnode_put(dst_vp); + fp_drop(p, fd2, fp2, 0); + error = EBADF; + goto outdrop; + } + + 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); + error = EBADF; + goto outdrop; + } + + /* Verify that both vps point to files and not directories */ + if ( !vnode_isreg(src_vp) || !vnode_isreg(dst_vp)) { + error = EINVAL; + vnode_put (src_vp); + vnode_put (dst_vp); + fp_drop (p, fd2, fp2, 0); + goto outdrop; + } + + /* + * The exchangedata syscall handler passes in 0 for the flags to VNOP_EXCHANGE. + * We'll pass in our special bit indicating that the new behavior is expected + */ + + error = VNOP_EXCHANGE(src_vp, dst_vp, FSOPT_EXCHANGE_DATA_ONLY, &context); + + vnode_put (src_vp); + vnode_put (dst_vp); + fp_drop(p, fd2, fp2, 0); + break; + } + + /* + * SPI for making a file compressed. + */ + case F_MAKECOMPRESSED: { + uint32_t gcounter = CAST_DOWN_EXPLICIT(uint32_t, uap->arg); + + if (fp->f_type != DTYPE_VNODE) { + error = EBADF; + goto out; + } + + vp = (struct vnode*) fp->f_data; + proc_fdunlock (p); + + /* get the vnode */ + if (vnode_getwithref(vp)) { + error = ENOENT; + goto outdrop; + } + + /* Is it a file? */ + if ((vnode_isreg(vp) == 0) && (vnode_islnk(vp) == 0)) { + vnode_put(vp); + error = EBADF; + goto outdrop; + } + + /* 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) { + vnode_put(vp); + error = EBADF; + goto outdrop; + } + + error = VNOP_IOCTL(vp, uap->cmd, (caddr_t)&gcounter, 0, &context); + + vnode_put (vp); + break; + } + + /* + * SPI (private) for indicating to a filesystem that subsequent writes to + * 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. + */ + + /* + * SPI (private) for indicating to a filesystem that subsequent writes to + * the open FD will represent static content. + */ + case F_SETSTATICCONTENT: { + caddr_t ioctl_arg = NULL; + + if (uap->arg) { + ioctl_arg = (caddr_t) 1; + } + + 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) { + error = ENOENT; + goto outdrop; + } + + /* 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) { + vnode_put(vp); + error = EBADF; + goto outdrop; + } + + error = VNOP_IOCTL(vp, uap->cmd, ioctl_arg, 0, &context); + (void)vnode_put(vp); + + break; + } + + /* + * SPI (private) for indicating to the lower level storage driver that the + * subsequent writes should be of a particular IO type (burst, greedy, static), + * or other flavors that may be necessary. + */ + case F_SETIOTYPE: { + caddr_t param_ptr; + uint32_t param; + + if (uap->arg) { + /* extract 32 bits of flags from userland */ + param_ptr = (caddr_t) uap->arg; + param = (uint32_t) param_ptr; + } + else { + /* If no argument is specified, error out */ + error = EINVAL; + goto out; + } + + /* + * Validate the different types of flags that can be specified: + * all of them are mutually exclusive for now. + */ + switch (param) { + case F_IOTYPE_ISOCHRONOUS: + break; + + default: + error = EINVAL; + goto out; + } + + + 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) { + error = ENOENT; + goto outdrop; + } + + /* 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) { + vnode_put(vp); + error = EBADF; + goto outdrop; + } + + error = VNOP_IOCTL(vp, uap->cmd, param_ptr, 0, &context); + (void)vnode_put(vp); + + break; + } + + + /* + * Extract the CodeDirectory of the vnode associated with + * the file descriptor and copy it back to user space + */ + case F_GETCODEDIR: { + struct user_fcodeblobs args; + + if (fp->f_type != DTYPE_VNODE) { + error = EBADF; + goto out; + } + + vp = (struct vnode *)fp->f_data; + proc_fdunlock(p); + + if ((fp->f_flag & FREAD) == 0) { + error = EBADF; + goto outdrop; + } + + if (IS_64BIT_PROCESS(p)) { + struct user64_fcodeblobs args64; + + error = copyin(argp, &args64, sizeof(args64)); + if (error) + goto outdrop; + + args.f_cd_hash = args64.f_cd_hash; + args.f_hash_size = args64.f_hash_size; + args.f_cd_buffer = args64.f_cd_buffer; + args.f_cd_size = args64.f_cd_size; + args.f_out_size = args64.f_out_size; + args.f_arch = args64.f_arch; + } else { + struct user32_fcodeblobs args32; + + error = copyin(argp, &args32, sizeof(args32)); + if (error) + goto outdrop; + + args.f_cd_hash = CAST_USER_ADDR_T(args32.f_cd_hash); + args.f_hash_size = args32.f_hash_size; + args.f_cd_buffer = CAST_USER_ADDR_T(args32.f_cd_buffer); + args.f_cd_size = args32.f_cd_size; + args.f_out_size = CAST_USER_ADDR_T(args32.f_out_size); + args.f_arch = args32.f_arch; + } + + if (vp->v_ubcinfo == NULL) { + error = EINVAL; + goto outdrop; + } + + struct cs_blob *t_blob = vp->v_ubcinfo->cs_blobs; + + /* + * This call fails if there is no cs_blob corresponding to the + * vnode, or if there are multiple cs_blobs present, and the caller + * did not specify which cpu_type they want the cs_blob for + */ + if (t_blob == NULL) { + error = ENOENT; /* there is no codesigning blob for this process */ + goto outdrop; + } else if (args.f_arch == 0 && t_blob->csb_next != NULL) { + error = ENOENT; /* too many architectures and none specified */ + goto outdrop; + } + + /* If the user specified an architecture, find the right blob */ + if (args.f_arch != 0) { + while (t_blob) { + if (t_blob->csb_cpu_type == args.f_arch) + break; + t_blob = t_blob->csb_next; + } + /* The cpu_type the user requested could not be found */ + if (t_blob == NULL) { + error = ENOENT; + goto outdrop; + } + } + + const CS_SuperBlob *super_blob = (void *)t_blob->csb_mem_kaddr; + const CS_CodeDirectory *cd = findCodeDirectory(super_blob, + (const char *) super_blob, + (const char *) super_blob + t_blob->csb_mem_size); + if (cd == NULL) { + error = ENOENT; + goto outdrop; + } + + uint64_t buffer_size = ntohl(cd->length); + + if (buffer_size > UINT_MAX) { + error = ERANGE; + goto outdrop; + } + + error = copyout(&buffer_size, args.f_out_size, sizeof(unsigned int)); + if (error) + goto outdrop; + + if (sizeof(t_blob->csb_cdhash) > args.f_hash_size || + buffer_size > args.f_cd_size) { + error = ERANGE; + goto outdrop; + } + + error = copyout(t_blob->csb_cdhash, args.f_cd_hash, sizeof(t_blob->csb_cdhash)); + if (error) + goto outdrop; + error = copyout(cd, args.f_cd_buffer, buffer_size); + if (error) + goto outdrop; + + break; + } + + /* + * Set the vnode pointed to by 'fd' + * and tag it as the (potentially future) backing store + * for another filesystem + */ + case F_SETBACKINGSTORE: { + if (fp->f_type != DTYPE_VNODE) { + error = EBADF; + goto out; + } + + vp = (struct vnode *)fp->f_data; + + if (vp->v_tag != VT_HFS) { + error = EINVAL; + goto out; + } + proc_fdunlock(p); + + if (vnode_getwithref(vp)) { + error = ENOENT; + goto outdrop; + } + + /* 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) { + vnode_put(vp); + error = EBADF; + goto outdrop; + } + + + /* 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); + } + + vnode_put(vp); + break; + } + + /* + * like F_GETPATH, but special semantics for + * the mobile time machine handler. + */ + case F_GETPATH_MTMINFO: { + char *pathbufp; + int pathlen; + + if (fp->f_type != DTYPE_VNODE) { + error = EBADF; + goto out; + } + vp = (struct vnode *)fp->f_data; + 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 ) { + 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); + } + (void)vnode_put(vp); + + if (error == 0) { + error = copyout((caddr_t)pathbufp, argp, pathlen); + } + if (error == 0) { + /* + * If the copyout was successful, now check to ensure + * that this vnode is not a BACKINGSTORE vnode. mtmd + * wants the path regardless. + */ + if (backingstore) { + error = EBUSY; + } + } + } else + (void)vnode_put(vp); + } + FREE(pathbufp, M_TEMP); + goto outdrop; + } + +#if DEBUG || DEVELOPMENT + case F_RECYCLE: + if (fp->f_type != DTYPE_VNODE) { + error = EBADF; + goto out; + } + vp = (struct vnode *)fp->f_data; + proc_fdunlock(p); + + vnode_recycle(vp); + break; +#endif + + default: + /* + * This is an fcntl() that we d not recognize at this level; + * if this is a vnode, we send it down into the VNOP_IOCTL + * 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; + goto out; + } + + /* Catch any now-invalid fcntl() selectors */ + switch (uap->cmd) { + case F_MARKDEPENDENCY: + error = EINVAL; + goto out; + default: + break; + } + + 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 ) { +#define STK_PARAMS 128 + char stkbuf[STK_PARAMS]; + unsigned int size; + caddr_t data, memp; + /* + * For this to work properly, we have to copy in the + * ioctl() cmd argument if there is one; we must also + * check that a command parameter, if present, does + * not exceed the maximum command length dictated by + * the number of bits we have available in the command + * to represent a structure length. Finally, we have + * to copy the results back out, if it is that type of + * ioctl(). + */ + size = IOCPARM_LEN(uap->cmd); + if (size > IOCPARM_MAX) { + (void)vnode_put(vp); + error = EINVAL; + break; + } + + memp = NULL; + if (size > sizeof (stkbuf)) { + if ((memp = (caddr_t)kalloc(size)) == 0) { + (void)vnode_put(vp); + error = ENOMEM; + goto outdrop; + } + data = memp; + } else { + data = &stkbuf[0]; + } + + if (uap->cmd & IOC_IN) { + if (size) { + /* structure */ + error = copyin(argp, data, size); + if (error) { + (void)vnode_put(vp); + if (memp) + kfree(memp, size); + goto outdrop; + } + + /* Bzero the section beyond that which was needed */ + if (size <= sizeof(stkbuf)) { + bzero ( (((uint8_t*)data) + size), (sizeof(stkbuf) - size)); + } + } else { + /* int */ + if (is64bit) { + *(user_addr_t *)data = argp; + } else { + *(uint32_t *)data = (uint32_t)argp; + } + }; + } else if ((uap->cmd & IOC_OUT) && size) { + /* + * Zero the buffer so the user always + * gets back something deterministic. + */ + bzero(data, size); + } else if (uap->cmd & IOC_VOID) { + if (is64bit) { + *(user_addr_t *)data = argp; + } else { + *(uint32_t *)data = (uint32_t)argp; + } + } + + error = VNOP_IOCTL(vp, uap->cmd, CAST_DOWN(caddr_t, data), 0, &context); + + (void)vnode_put(vp); + + /* Copy any output data to user */ + if (error == 0 && (uap->cmd & IOC_OUT) && size) + error = copyout(data, argp, size); + if (memp) + kfree(memp, size); + } + break; + } + +outdrop: + AUDIT_ARG(vnpath_withref, vp, ARG_VNODE1); + fp_drop(p, fd, fp, 0); + return(error); +out: + fp_drop(p, fd, fp, 1); + proc_fdunlock(p); + return(error); +} + + +/* + * finishdup + * + * Description: Common code for dup, dup2, and fcntl(F_DUPFD). + * + * Parameters: p Process performing the dup + * old The fd to dup + * new The fd to dup it to + * fd_flags Flags to augment the new fd + * retval Pointer to the call return area + * + * Returns: 0 Success + * EBADF + * ENOMEM + * + * Implicit returns: + * *retval (modified) The new descriptor + * + * Locks: Assumes proc_fdlock for process pointing to fdp is held by + * the caller + * + * 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 +finishdup(proc_t p, + struct filedesc *fdp, int old, int new, int fd_flags, int32_t *retval) +{ + struct fileproc *nfp; + struct fileproc *ofp; +#if CONFIG_MACF + int error; +#endif + +#if DIAGNOSTIC + proc_fdlock_assert(p, LCK_MTX_ASSERT_OWNED); +#endif + if ((ofp = fdp->fd_ofiles[old]) == NULL || + (fdp->fd_ofileflags[old] & UF_RESERVED)) { + fdrelse(p, new); + return (EBADF); + } + fg_ref(ofp); + +#if CONFIG_MACF + error = mac_file_check_dup(proc_ucred(p), ofp->f_fglob, new); + if (error) { + fg_drop(ofp); + fdrelse(p, new); + return (error); + } +#endif + + proc_fdunlock(p); + + nfp = fileproc_alloc_init(NULL); + + proc_fdlock(p); + + if (nfp == NULL) { + fg_drop(ofp); + fdrelse(p, new); + return (ENOMEM); + } + + nfp->f_fglob = ofp->f_fglob; + +#if DIAGNOSTIC + if (fdp->fd_ofiles[new] != 0) + panic("finishdup: overwriting fd_ofiles with new %d", new); + if ((fdp->fd_ofileflags[new] & UF_RESERVED) == 0) + panic("finishdup: unreserved fileflags with new %d", new); +#endif + + if (new > fdp->fd_lastfile) + fdp->fd_lastfile = new; + *fdflags(p, new) |= fd_flags; + procfdtbl_releasefd(p, new, nfp); + *retval = new; + return (0); +} + + +/* + * close + * + * Description: The implementation of the close(2) system call + * + * Parameters: p Process in whose per process file table + * the close is to occur + * uap->fd fd to be closed + * retval + * + * Returns: 0 Success + * fp_lookup:EBADF Bad file descriptor + * fp_guard_exception:??? Guarded file descriptor + * close_internal:EBADF + * close_internal:??? Anything returnable by a per-fileops + * close function + */ +int +close(proc_t p, struct close_args *uap, int32_t *retval) +{ + __pthread_testcancel(1); + return(close_nocancel(p, (struct close_nocancel_args *)uap, retval)); +} + + +int +close_nocancel(proc_t p, struct close_nocancel_args *uap, __unused int32_t *retval) +{ + struct fileproc *fp; + int fd = uap->fd; + int error; + + AUDIT_SYSCLOSE(p, fd); + + proc_fdlock(p); + + if ( (error = fp_lookup(p,fd,&fp, 1)) ) { + proc_fdunlock(p); + return(error); + } + + if (FP_ISGUARDED(fp, GUARD_CLOSE)) { + error = fp_guard_exception(p, fd, fp, kGUARD_EXC_CLOSE); + (void) fp_drop(p, fd, fp, 1); + proc_fdunlock(p); + return (error); + } + + error = close_internal_locked(p, fd, fp, 0); + + proc_fdunlock(p); + + return (error); +} /* + * 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 -close_internal(struct proc *p, int fd, struct fileproc *fp, int flags) +close_internal_locked(proc_t p, int fd, struct fileproc *fp, int flags) { struct filedesc *fdp = p->p_fd; int error =0; - int locked = flags & CLOSEINT_LOCKED; - int waitonclose = flags & CLOSEINT_WAITONCLOSE; - int norelse = flags & CLOSEINT_NOFDRELSE; - int nofdref = flags & CLOSEINT_NOFDNOREF; - int slpstate = PRIBIO; + int resvfd = flags & FD_DUP2RESV; - if (!locked) - proc_fdlock(p); - /* Keep people from using the filedesc while we are closing it */ - fdp->fd_ofileflags[fd] |= UF_RESERVED; +#if DIAGNOSTIC + proc_fdlock_assert(p, LCK_MTX_ASSERT_OWNED); +#endif - fdp->fd_ofileflags[fd] |= UF_CLOSING; + /* Keep people from using the filedesc while we are closing it */ + procfdtbl_markclosefd(p, fd); - if ((waitonclose && ((fp->f_flags & FP_CLOSING) == FP_CLOSING))) { - if (nofdref == 0) - fp_drop(p, fd, fp, 1); - fp->f_flags |= FP_WAITCLOSE; - if (!locked) - slpstate |= PDROP; - msleep(&fp->f_flags, &p->p_fdmlock, slpstate, "close wait",0) ; - return(EBADF); + if ((fp->f_flags & FP_CLOSING) == FP_CLOSING) { + panic("close_internal_locked: being called on already closing fd"); } + +#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 (nofdref) - fp->f_iocount++; if ( (fp->f_flags & FP_AIOISSUED) || kauth_authorize_fileop_has_listeners() ) { @@ -1069,65 +2868,121 @@ close_internal(struct proc *p, int fd, struct fileproc *fp, int flags) if (fp->f_flags & FP_WAITEVENT) (void)waitevent_close(p, fp); - if ((fp->f_flags & FP_INCHRREAD) == 0) - fileproc_drain(p, fp); - if (norelse == 0) - _fdrelse(fdp, fd); + fileproc_drain(p, fp); + + if (resvfd == 0) { + _fdrelse(p, fd); + } else { + procfdtbl_reservefd(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); - if (!locked) - 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); + } +#endif - FREE_ZONE(fp, sizeof *fp, M_FILEPROC); return(error); } + /* - * Return status information about a file descriptor. + * fstat1 + * + * Description: Return status information about a file descriptor. + * + * Parameters: p The process doing the fstat + * fd The fd to stat + * ub The user stat buffer + * xsecurity The user extended security + * buffer, or 0 if none + * xsecurity_size The size of xsecurity, or 0 + * if no xsecurity + * isstat64 Flag to indicate 64 bit version + * for inode size, etc. + * + * Returns: 0 Success + * EBADF + * EFAULT + * fp_lookup:EBADF Bad file descriptor + * vnode_getwithref:??? + * copyout:EFAULT + * vnode_getwithref:??? + * vn_stat:??? + * soo_stat:??? + * pipe_stat:??? + * pshm_stat:??? + * kqueue_stat:??? * - * XXX switch on node type is bogus; need a stat in struct fileops instead. + * Notes: Internal implementation for all other fstat() related + * functions + * + * XXX switch on node type is bogus; need a stat in struct + * XXX fileops instead. */ static int -fstat1(struct proc *p, int fd, user_addr_t ub, user_addr_t xsecurity, user_addr_t xsecurity_size) +fstat1(proc_t p, int fd, user_addr_t ub, user_addr_t xsecurity, user_addr_t xsecurity_size, int isstat64) { struct fileproc *fp; - struct stat sb; - struct user_stat user_sb; + union { + struct stat sb; + struct stat64 sb64; + } source; + union { + struct user64_stat user64_sb; + struct user32_stat user32_sb; + struct user64_stat64 user64_sb64; + struct user32_stat64 user32_sb64; + } dest; int error, my_size; - int funnel_state; file_type_t type; caddr_t data; kauth_filesec_t fsec; - ssize_t xsecurity_bufsize; - int entrycount; - struct vfs_context context; + user_size_t xsecurity_bufsize; + vfs_context_t ctx = vfs_context_current(); + void * sbptr; AUDIT_ARG(fd, fd); - if ((error = fp_lookup(p, fd, &fp, 0)) != 0) + if ((error = fp_lookup(p, fd, &fp, 0)) != 0) { return(error); + } type = fp->f_type; data = fp->f_data; fsec = KAUTH_FILESEC_NONE; + sbptr = (void *)&source; + switch (type) { case DTYPE_VNODE: - context.vc_proc = current_proc(); - context.vc_ucred = kauth_cred_get(); if ((error = vnode_getwithref((vnode_t)data)) == 0) { /* - * If the caller has the file open, and is not requesting extended security, - * we are going to let them get the basic stat information. + * If the caller has the file open, and is not + * requesting extended security information, we are + * going to let them get the basic stat information. */ if (xsecurity == USER_ADDR_NULL) { - error = vn_stat_noauth((vnode_t)data, &sb, NULL, &context); + error = vn_stat_noauth((vnode_t)data, sbptr, NULL, isstat64, ctx); } else { - error = vn_stat((vnode_t)data, &sb, &fsec, &context); + error = vn_stat((vnode_t)data, sbptr, &fsec, isstat64, ctx); } AUDIT_ARG(vnpath, (struct vnode *)data, ARG_VNODE1); @@ -1135,43 +2990,60 @@ fstat1(struct proc *p, int fd, user_addr_t ub, user_addr_t xsecurity, user_addr_ } break; +#if SOCKETS case DTYPE_SOCKET: - error = soo_stat((struct socket *)data, &sb); + error = soo_stat((struct socket *)data, sbptr, isstat64); break; +#endif /* SOCKETS */ case DTYPE_PIPE: - error = pipe_stat((void *)data, &sb); + error = pipe_stat((void *)data, sbptr, isstat64); break; case DTYPE_PSXSHM: - error = pshm_stat((void *)data, &sb); + error = pshm_stat((void *)data, sbptr, isstat64); break; case DTYPE_KQUEUE: - funnel_state = thread_funnel_set(kernel_flock, TRUE); - error = kqueue_stat(fp, &sb, p); - thread_funnel_set(kernel_flock, funnel_state); + error = kqueue_stat((void *)data, sbptr, isstat64, p); break; default: error = EBADF; goto out; } - /* Zap spare fields */ - sb.st_lspare = 0; - sb.st_qspare[0] = 0LL; - sb.st_qspare[1] = 0LL; if (error == 0) { caddr_t sbp; - if (IS_64BIT_PROCESS(current_proc())) { - munge_stat(&sb, &user_sb); - my_size = sizeof(user_sb); - sbp = (caddr_t)&user_sb; - } - else { - my_size = sizeof(sb); - sbp = (caddr_t)&sb; + + if (isstat64 != 0) { + source.sb64.st_lspare = 0; + source.sb64.st_qspare[0] = 0LL; + source.sb64.st_qspare[1] = 0LL; + + if (IS_64BIT_PROCESS(current_proc())) { + munge_user64_stat64(&source.sb64, &dest.user64_sb64); + my_size = sizeof(dest.user64_sb64); + sbp = (caddr_t)&dest.user64_sb64; + } else { + munge_user32_stat64(&source.sb64, &dest.user32_sb64); + my_size = sizeof(dest.user32_sb64); + sbp = (caddr_t)&dest.user32_sb64; + } + } else { + source.sb.st_lspare = 0; + source.sb.st_qspare[0] = 0LL; + source.sb.st_qspare[1] = 0LL; + if (IS_64BIT_PROCESS(current_proc())) { + munge_user64_stat(&source.sb, &dest.user64_sb); + my_size = sizeof(dest.user64_sb); + sbp = (caddr_t)&dest.user64_sb; + } else { + munge_user32_stat(&source.sb, &dest.user32_sb); + my_size = sizeof(dest.user32_sb); + sbp = (caddr_t)&dest.user32_sb; + } } + error = copyout(sbp, ub, my_size); } @@ -1206,31 +3078,117 @@ out: return (error); } + +/* + * fstat_extended + * + * Description: Extended version of fstat supporting returning extended + * security information + * + * Parameters: p The process doing the fstat + * uap->fd The fd to stat + * uap->ub The user stat buffer + * uap->xsecurity The user extended security + * buffer, or 0 if none + * uap->xsecurity_size The size of xsecurity, or 0 + * + * Returns: 0 Success + * !0 Errno (see fstat1) + */ +int +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)); +} + + +/* + * fstat + * + * Description: Get file status for the file associated with fd + * + * Parameters: p The process doing the fstat + * uap->fd The fd to stat + * uap->ub The user stat buffer + * + * Returns: 0 Success + * !0 Errno (see fstat1) + */ +int +fstat(proc_t p, register struct fstat_args *uap, __unused int32_t *retval) +{ + return(fstat1(p, uap->fd, uap->ub, 0, 0, 0)); +} + + +/* + * fstat64_extended + * + * Description: Extended version of fstat64 supporting returning extended + * security information + * + * Parameters: p The process doing the fstat + * uap->fd The fd to stat + * uap->ub The user stat buffer + * uap->xsecurity The user extended security + * buffer, or 0 if none + * uap->xsecurity_size The size of xsecurity, or 0 + * + * Returns: 0 Success + * !0 Errno (see fstat1) + */ int -fstat_extended(struct proc *p, struct fstat_extended_args *uap, __unused register_t *retval) +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)); + return(fstat1(p, uap->fd, uap->ub, uap->xsecurity, uap->xsecurity_size, 1)); } + +/* + * fstat64 + * + * Description: Get 64 bit version of the file status for the file associated + * with fd + * + * Parameters: p The process doing the fstat + * uap->fd The fd to stat + * uap->ub The user stat buffer + * + * Returns: 0 Success + * !0 Errno (see fstat1) + */ int -fstat(struct proc *p, register struct fstat_args *uap, __unused register_t *retval) +fstat64(proc_t p, register struct fstat64_args *uap, __unused int32_t *retval) { - return(fstat1(p, uap->fd, uap->ub, 0, 0)); + return(fstat1(p, uap->fd, uap->ub, 0, 0, 1)); } + /* - * Return pathconf information about a file descriptor. + * fpathconf + * + * Description: Return pathconf information about a file descriptor. + * + * Parameters: p Process making the request + * uap->fd fd to get information about + * uap->name Name of information desired + * retval Pointer to the call return area + * + * Returns: 0 Success + * EINVAL + * fp_lookup:EBADF Bad file descriptor + * vnode_getwithref:??? + * vn_pathconf:??? + * + * Implicit returns: + * *retval (modified) Returned information (numeric) */ int -fpathconf(p, uap, retval) - struct proc *p; - register struct fpathconf_args *uap; - register_t *retval; +fpathconf(proc_t p, struct fpathconf_args *uap, int32_t *retval) { int fd = uap->fd; struct fileproc *fp; struct vnode *vp; - struct vfs_context context; int error = 0; file_type_t type; caddr_t data; @@ -1245,8 +3203,8 @@ fpathconf(p, uap, retval) switch (type) { case DTYPE_SOCKET: - if (uap->name != _PC_PIPE_BUF) { - error = EINVAL; + if (uap->name != _PC_PIPE_BUF) { + error = EINVAL; goto out; } *retval = PIPE_BUF; @@ -1254,7 +3212,11 @@ fpathconf(p, uap, retval) goto out; case DTYPE_PIPE: - *retval = PIPE_BUF; + if (uap->name != _PC_PIPE_BUF) { + error = EINVAL; + goto out; + } + *retval = PIPE_BUF; error = 0; goto out; @@ -1264,19 +3226,13 @@ fpathconf(p, uap, retval) if ( (error = vnode_getwithref(vp)) == 0) { AUDIT_ARG(vnpath, vp, ARG_VNODE1); - context.vc_proc = p; - context.vc_ucred = kauth_cred_get(); - - error = vn_pathconf(vp, uap->name, retval, &context); + error = vn_pathconf(vp, uap->name, retval, vfs_context_current()); (void)vnode_put(vp); } goto out; - case DTYPE_PSXSHM: - case DTYPE_PSXSEM: - case DTYPE_KQUEUE: - case DTYPE_FSEVENTS: + default: error = EINVAL; goto out; @@ -1288,37 +3244,56 @@ out: } /* - * Allocate a file descriptor for the process. + * Statistics counter for the number of times a process calling fdalloc() + * has resulted in an expansion of the per process open file table. + * + * XXX This would likely be of more use if it were per process + */ +int fdexpand; + + +/* + * fdalloc + * + * Description: Allocate a file descriptor for the process. + * + * Parameters: p Process to allocate the fd in + * want The fd we would prefer to get + * result Pointer to fd we got + * + * Returns: 0 Success + * EMFILE + * ENOMEM + * + * Implicit returns: + * *result (modified) The fd which was allocated */ -int fdexpand; - int -fdalloc(p, want, result) - struct proc *p; - int want; - int *result; +fdalloc(proc_t p, int want, int *result) { - register struct filedesc *fdp = p->p_fd; - register int i; + struct filedesc *fdp = p->p_fd; + int i; int lim, last, numfiles, oldnfiles; struct fileproc **newofiles, **ofiles; - char *newofileflags, *ofileflags; + char *newofileflags; /* * Search for a free descriptor starting at the higher * of want or fd_freefile. If that fails, consider * expanding the ofile array. */ +#if DIAGNOSTIC + proc_fdlock_assert(p, LCK_MTX_ASSERT_OWNED); +#endif + lim = min((int)p->p_rlimit[RLIMIT_NOFILE].rlim_cur, maxfiles); for (;;) { last = min(fdp->fd_nfiles, lim); if ((i = want) < fdp->fd_freefile) i = fdp->fd_freefile; - ofiles = &fdp->fd_ofiles[i]; - ofileflags = &fdp->fd_ofileflags[i]; for (; i < last; i++) { - if (*ofiles == NULL && !(*ofileflags & UF_RESERVED)) { - *ofileflags = UF_RESERVED; + if (fdp->fd_ofiles[i] == NULL && !(fdp->fd_ofileflags[i] & UF_RESERVED)) { + procfdtbl_reservefd(p, i); if (i > fdp->fd_lastfile) fdp->fd_lastfile = i; if (want <= fdp->fd_freefile) @@ -1326,7 +3301,6 @@ fdalloc(p, want, result) *result = i; return (0); } - ofiles++; ofileflags++; } /* @@ -1359,15 +3333,15 @@ fdalloc(p, want, result) */ 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, - (numfiles - oldnfiles) * sizeof *fdp->fd_ofiles); + (numfiles - oldnfiles) * sizeof(*fdp->fd_ofiles)); (void) memcpy(newofileflags, fdp->fd_ofileflags, - oldnfiles * sizeof *fdp->fd_ofileflags); + oldnfiles * sizeof(*fdp->fd_ofileflags)); (void) memset(&newofileflags[oldnfiles], 0, (numfiles - oldnfiles) * - sizeof *fdp->fd_ofileflags); + sizeof(*fdp->fd_ofileflags)); ofiles = fdp->fd_ofiles; fdp->fd_ofiles = newofiles; fdp->fd_ofileflags = newofileflags; @@ -1377,14 +3351,26 @@ fdalloc(p, want, result) } } + /* - * Check to see whether n user file descriptors - * are available to the process p. + * fdavail + * + * Description: Check to see whether n user file descriptors are available + * to the process p. + * + * Parameters: p Process to check in + * n The number of fd's desired + * + * Returns: 0 No + * 1 Yes + * + * Locks: Assumes proc_fdlock for process is held by the caller + * + * Notes: The answer only remains valid so long as the proc_fdlock is + * held by the caller. */ int -fdavail(p, n) - struct proc *p; - int n; +fdavail(proc_t p, int n) { struct filedesc *fdp = p->p_fd; struct fileproc **fpp; @@ -1402,45 +3388,50 @@ fdavail(p, n) return (0); } -void -fdrelse(p, fd) - struct proc *p; - int fd; -{ - _fdrelse(p->p_fd, fd); -} +/* + * 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 -fddrop(p, fd) - struct proc *p; - int fd; +fdrelse(proc_t p, int fd) { - struct filedesc *fdp = p->p_fd; - struct fileproc *fp; - - if (fd < fdp->fd_freefile) - fdp->fd_freefile = fd; -#if DIAGNOSTIC - if (fd > fdp->fd_lastfile) - panic("fdrelse: fd_lastfile inconsistent"); -#endif - fp = fdp->fd_ofiles[fd]; - fdp->fd_ofiles[fd] = NULL; - fdp->fd_ofileflags[fd] = 0; - - while ((fd = fdp->fd_lastfile) > 0 && - fdp->fd_ofiles[fd] == NULL && - !(fdp->fd_ofileflags[fd] & UF_RESERVED)) - fdp->fd_lastfile--; - FREE_ZONE(fp, sizeof *fp, M_FILEPROC); + _fdrelse(p, fd); } +/* + * 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 -fdgetf_noref(p, fd, resultfp) - struct proc *p; - int fd; - struct fileproc **resultfp; +fdgetf_noref(proc_t p, int fd, struct fileproc **resultfp) { struct filedesc *fdp = p->p_fd; struct fileproc *fp; @@ -1456,36 +3447,99 @@ fdgetf_noref(p, fd, resultfp) } -/* should be called only when proc_fdlock is held */ -void -fp_setflags(proc_t p, struct fileproc * fp, int flags) +/* + * fp_getfvp + * + * 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 + * + * 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 + * + * Returns: 0 Success + * EBADF Bad file descriptor + * ENOTSUP fd does not refer to a vnode + * + * Implicit returns: + * *resultfp (modified) Fileproc pointer + * *resultvp (modified) vnode pointer + * + * Notes: The resultfp and resultvp fields are optional, and may be + * independently specified as NULL to skip returning information + * + * Locks: Internally takes and releases proc_fdlock + */ +int +fp_getfvp(proc_t p, int fd, struct fileproc **resultfp, struct vnode **resultvp) { - proc_fdlock(p); - fp->f_flags |= flags; - proc_fdunlock(p); -} + struct filedesc *fdp = p->p_fd; + struct fileproc *fp; -void -fp_clearflags(proc_t p, struct fileproc * fp, int flags) -{ + 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); + } + fp->f_iocount++; - proc_fdlock(p); - if (fp) - fp->f_flags &= ~flags; + if (resultfp) + *resultfp = fp; + if (resultvp) + *resultvp = (struct vnode *)fp->f_data; proc_fdunlock(p); + + return (0); } + +/* + * fp_getfvpandvid + * + * 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 + * + * 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 + * + * Returns: 0 Success + * EBADF Bad file descriptor + * ENOTSUP fd does not refer to a vnode + * + * 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 + * + * Locks: Internally takes and releases proc_fdlock + */ int -fp_getfvp(p, fd, resultfp, resultvp) - struct proc *p; - int fd; - struct fileproc **resultfp; - struct vnode **resultvp; +fp_getfvpandvid(proc_t p, int fd, struct fileproc **resultfp, + struct vnode **resultvp, uint32_t *vidp) { struct filedesc *fdp = p->p_fd; struct fileproc *fp; - proc_fdlock(p); + proc_fdlock_spin(p); if (fd < 0 || fd >= fdp->fd_nfiles || (fp = fdp->fd_ofiles[fd]) == NULL || (fdp->fd_ofileflags[fd] & UF_RESERVED)) { @@ -1502,6 +3556,8 @@ fp_getfvp(p, fd, resultfp, resultvp) *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); return (0); @@ -1509,24 +3565,38 @@ fp_getfvp(p, fd, resultfp, resultvp) /* + * fp_getfsock + * + * 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 + * + * 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 + * * Returns: EBADF The file descriptor is invalid * EOPNOTSUPP The file descriptor is not a socket * 0 Success * + * Implicit returns: + * *resultfp (modified) Fileproc pointer + * *results (modified) socket pointer + * * Notes: EOPNOTSUPP should probably be ENOTSOCK; this function is only * ever called from accept1(). */ int -fp_getfsock(p, fd, resultfp, results) - struct proc *p; - int fd; - struct fileproc **resultfp; - struct socket **results; +fp_getfsock(proc_t p, int fd, struct fileproc **resultfp, + struct socket **results) { struct filedesc *fdp = p->p_fd; struct fileproc *fp; - proc_fdlock(p); + proc_fdlock_spin(p); if (fd < 0 || fd >= fdp->fd_nfiles || (fp = fdp->fd_ofiles[fd]) == NULL || (fdp->fd_ofileflags[fd] & UF_RESERVED)) { @@ -1549,17 +3619,39 @@ fp_getfsock(p, fd, resultfp, results) } +/* + * 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(p, fd, resultfp, resultkq) - struct proc *p; - int fd; - struct fileproc **resultfp; - struct kqueue **resultkq; +fp_getfkq(proc_t p, int fd, struct fileproc **resultfp, + struct kqueue **resultkq) { struct filedesc *fdp = p->p_fd; struct fileproc *fp; - proc_fdlock(p); + proc_fdlock_spin(p); if ( fd < 0 || fd >= fdp->fd_nfiles || (fp = fdp->fd_ofiles[fd]) == NULL || (fdp->fd_ofileflags[fd] & UF_RESERVED)) { @@ -1581,17 +3673,42 @@ fp_getfkq(p, fd, resultfp, resultkq) 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(p, fd, resultfp, resultpshm) - struct proc *p; - int fd; - struct fileproc **resultfp; - struct pshmnode **resultpshm; +fp_getfpshm(proc_t p, int fd, struct fileproc **resultfp, + struct pshmnode **resultpshm) { struct filedesc *fdp = p->p_fd; struct fileproc *fp; - proc_fdlock(p); + proc_fdlock_spin(p); if (fd < 0 || fd >= fdp->fd_nfiles || (fp = fdp->fd_ofiles[fd]) == NULL || (fdp->fd_ofileflags[fd] & UF_RESERVED)) { @@ -1615,17 +3732,50 @@ fp_getfpshm(p, fd, resultfp, resultpshm) } +/* + * 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(p, fd, resultfp, resultpsem) - struct proc *p; - int fd; - struct fileproc **resultfp; - struct psemnode **resultpsem; +fp_getfpsem(proc_t p, int fd, struct fileproc **resultfp, + struct psemnode **resultpsem) { struct filedesc *fdp = p->p_fd; struct fileproc *fp; - proc_fdlock(p); + proc_fdlock_spin(p); if (fd < 0 || fd >= fdp->fd_nfiles || (fp = fdp->fd_ofiles[fd]) == NULL || (fdp->fd_ofileflags[fd] & UF_RESERVED)) { @@ -1646,21 +3796,97 @@ fp_getfpsem(p, fd, resultfp, resultpsem) 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_lookup(p, fd, resultfp, locked) - struct proc *p; - int fd; - struct fileproc **resultfp; - int locked; +fp_getfpipe(proc_t p, int fd, struct fileproc **resultfp, + struct pipe **resultpipe) { struct filedesc *fdp = p->p_fd; struct fileproc *fp; - if (!locked) - proc_fdlock(p); + 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); @@ -1675,12 +3901,98 @@ fp_lookup(p, fd, resultfp, locked) return (0); } + +/* + * 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. + * + * Parameters: p Process containing the fd + * fd The fd of interest + * nfp Pointer to the newfp + * + * Returns: 0 Success + * EBADF Bad file descriptor + * EINTR Interrupted + * EKEEPLOOKING f_iocount changed while lock was dropped. + */ +int +fp_tryswap(proc_t p, int fd, struct fileproc *nfp) +{ + struct fileproc *fp; + int error; + + proc_fdlock_assert(p, LCK_MTX_ASSERT_OWNED); + + if (0 != (error = fp_lookup(p, fd, &fp, 1))) + return (error); + /* + * At this point, our caller (change_guardedfd_np) has + * one f_iocount reference, and we just took another + * one to begin the replacement. + */ + if (fp->f_iocount < 2) { + panic("f_iocount too small %d", fp->f_iocount); + } else if (2 == fp->f_iocount) { + + /* 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; + + p->p_fd->fd_ofiles[fd] = nfp; + (void) fp_drop(p, fd, nfp, 1); + } else { + /* + * Wait for all other active references to evaporate. + */ + p->p_fpdrainwait = 1; + error = msleep(&p->p_fpdrainwait, &p->p_fdmlock, + PRIBIO | PCATCH, "tryswap fpdrain", NULL); + if (0 == error) { + /* + * Return an "internal" errno to trigger a full + * reevaluation of the change-guard attempt. + */ + error = EKEEPLOOKING; + printf("%s: lookup collision fd %d\n", __func__, fd); + } + (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(p); + proc_fdlock_spin(p); fp->f_flags |= FP_WRITTEN; @@ -1692,12 +4004,31 @@ fp_drop_written(proc_t p, int fd, struct fileproc *fp) } +/* + * 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(p); + proc_fdlock_spin(p); fp->f_flags |= FP_WAITEVENT; @@ -1708,17 +4039,38 @@ fp_drop_event(proc_t p, int fd, struct fileproc *fp) return (error); } + +/* + * fp_drop + * + * Description: 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 + * locked flag to internally take and + * drop proc_fdlock if it is not + * already held by the caller + * + * Returns: 0 Success + * EBADF Bad file descriptor + * + * Locks: This function internally takes and drops the proc_fdlock for + * the supplied process if 'locked' is non-zero, and assumes that + * the caller already holds this lock if 'locked' is non-zero. + * + * Notes: The fileproc must correspond to the fd in the supplied proc + */ int -fp_drop(p, fd, fp, locked) - struct proc *p; - int fd; - struct fileproc *fp; - int locked; +fp_drop(proc_t p, int fd, struct fileproc *fp, int locked) { struct filedesc *fdp = p->p_fd; + int needwakeup = 0; if (!locked) - proc_fdlock(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) && @@ -1729,24 +4081,129 @@ fp_drop(p, fd, fp, locked) } fp->f_iocount--; - if (p->p_fpdrainwait && fp->f_iocount == 0) { - p->p_fpdrainwait = 0; + if (fp->f_iocount == 0) { + if (fp->f_flags & FP_SELCONFLICT) + fp->f_flags &= ~FP_SELCONFLICT; + + if (p->p_fpdrainwait) { + p->p_fpdrainwait = 0; + needwakeup = 1; + } + } + if (!locked) + proc_fdunlock(p); + if (needwakeup) wakeup(&p->p_fpdrainwait); + + return (0); +} + + +/* + * file_vnode + * + * Description: Given an fd, look it up in the current process's per process + * open file table, and return its internal vnode pointer. + * + * Parameters: fd fd to obtain vnode from + * vpp pointer to vnode return area + * + * Returns: 0 Success + * EINVAL The fd does not refer to a + * vnode fileproc entry + * fp_lookup:EBADF Bad file descriptor + * + * Implicit returns: + * *vpp (modified) Returned vnode pointer + * + * 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 + * fd's corresponding fileproc. + * + * The fileproc referenced is not returned; because of this, care + * must be taken to not drop the last reference (e.g. by closing + * the file). This is inherently unsafe, since the reference may + * 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 + * can result in the returned pointer immediately becoming invalid + * following the call. + * + * Use of this function is discouraged. + */ +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 (!locked) + if (fp->f_type != DTYPE_VNODE) { + fp_drop(p, fd, fp,1); proc_fdunlock(p); - - return (0); + return(EINVAL); + } + if (vpp != NULL) + *vpp = (struct vnode *)fp->f_data; + proc_fdunlock(p); + + return(0); } + +/* + * file_vnode_withvid + * + * Description: Given an fd, look it up in the current process's per process + * open file table, and return its internal vnode pointer. + * + * Parameters: fd fd to obtain vnode from + * vpp pointer to vnode return area + * vidp pointer to vid of the returned vnode + * + * Returns: 0 Success + * EINVAL The fd does not refer to a + * vnode fileproc entry + * fp_lookup:EBADF Bad file descriptor + * + * Implicit returns: + * *vpp (modified) Returned vnode pointer + * + * 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 + * fd's corresponding fileproc. + * + * The fileproc referenced is not returned; because of this, care + * must be taken to not drop the last reference (e.g. by closing + * the file). This is inherently unsafe, since the reference may + * 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 + * can result in the returned pointer immediately becoming invalid + * following the call. + * + * Use of this function is discouraged. + */ int -file_vnode(int fd, struct vnode **vpp) +file_vnode_withvid(int fd, struct vnode **vpp, uint32_t * vidp) { - struct proc * p = current_proc(); + proc_t p = current_proc(); struct fileproc *fp; + vnode_t vp; int error; - proc_fdlock(p); + proc_fdlock_spin(p); if ( (error = fp_lookup(p, fd, &fp, 1)) ) { proc_fdunlock(p); return(error); @@ -1756,21 +4213,61 @@ file_vnode(int fd, struct vnode **vpp) proc_fdunlock(p); return(EINVAL); } - *vpp = (struct vnode *)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); } +/* + * file_socket + * + * Description: Given an fd, look it up in the current process's per process + * open file table, and return its internal socket pointer. + * + * Parameters: fd fd to obtain vnode from + * sp pointer to socket return area + * + * Returns: 0 Success + * ENOTSOCK Not a socket + * fp_lookup:EBADF Bad file descriptor + * + * Implicit returns: + * *sp (modified) Returned socket pointer + * + * 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 + * fd's corresponding fileproc. + * + * The fileproc referenced is not returned; because of this, care + * must be taken to not drop the last reference (e.g. by closing + * the file). This is inherently unsafe, since the reference may + * 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 + * can result in the returned pointer immediately becoming invalid + * following the call. + * + * Use of this function is discouraged. + */ int file_socket(int fd, struct socket **sp) { - struct proc * p = current_proc(); + proc_t p = current_proc(); struct fileproc *fp; int error; - proc_fdlock(p); + proc_fdlock_spin(p); if ( (error = fp_lookup(p, fd, &fp, 1)) ) { proc_fdunlock(p); return(error); @@ -1786,15 +4283,39 @@ file_socket(int fd, struct socket **sp) return(0); } + +/* + * file_flags + * + * Description: Given an fd, look it up in the current process's per process + * open file table, and return its fileproc's flags field. + * + * Parameters: fd fd whose flags are to be + * retrieved + * flags pointer to flags data area + * + * Returns: 0 Success + * ENOTSOCK Not a socket + * fp_lookup:EBADF Bad file descriptor + * + * Implicit returns: + * *flags (modified) Returned flags field + * + * 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) +file_flags(int fd, int *flags) { - struct proc * p = current_proc(); + proc_t p = current_proc(); struct fileproc *fp; int error; - proc_fdlock(p); + proc_fdlock_spin(p); if ( (error = fp_lookup(p, fd, &fp, 1)) ) { proc_fdunlock(p); return(error); @@ -1807,13 +4328,50 @@ file_flags(int fd, int * flags) } +/* + * file_drop + * + * Description: Drop an iocount reference on an fd, and wake up any waiters + * for draining (i.e. blocked in fileproc_drain() called during + * the last attempt to close a file). + * + * Parameters: fd fd on which an ioreference is + * to be dropped + * + * Returns: 0 Success + * EBADF Bad file descriptor + * + * 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 + * + * 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.). + * + * Use of this function is discouraged. + */ int file_drop(int fd) { struct fileproc *fp; - struct proc *p = current_proc(); + proc_t p = current_proc(); + int needwakeup = 0; - proc_fdlock(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) && @@ -1823,42 +4381,142 @@ file_drop(int fd) } fp->f_iocount --; - if (p->p_fpdrainwait && fp->f_iocount == 0) { - p->p_fpdrainwait = 0; - wakeup(&p->p_fpdrainwait); + if (fp->f_iocount == 0) { + if (fp->f_flags & FP_SELCONFLICT) + fp->f_flags &= ~FP_SELCONFLICT; + + if (p->p_fpdrainwait) { + p->p_fpdrainwait = 0; + needwakeup = 1; + } } proc_fdunlock(p); + + 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(p, resultfp, resultfd ) - struct proc *p; - struct fileproc **resultfp; - int *resultfd; +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_locked(p, resultfp, resultfd, 1); + error = falloc_withalloc_locked(p, + resultfp, resultfd, ctx, fp_zalloc, arg, 1); proc_fdunlock(p); - return(error); + return (error); } + +/* + * "uninitialized" ops -- ensure fg->fg_ops->fo_type always exists + */ +static const struct fileops uninitops; + /* + * falloc_locked + * * Create a new open file structure and allocate - * a file decriptor for the process that refers to it. + * a file descriptor for the process that refers to it. + * + * Returns: 0 Success + * + * 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 + * locked Flag to indicate whether the + * caller holds proc_fdlock + * + * 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 + * exhausted + * fdalloc:ENOMEM + * + * 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 + * solely to support kern_exec.c; otherwise, it would take + * neither, and use the vfs_context_current() routine internally. */ int -falloc_locked(p, resultfp, resultfd, locked) - struct proc *p; - struct fileproc **resultfp; - int *resultfd; - int locked; +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) { - struct fileproc *fp, *fq; + struct fileproc *fp; struct fileglob *fg; int error, nfd; @@ -1875,6 +4533,15 @@ falloc_locked(p, resultfp, resultfd, locked) tablefull("file"); return (ENFILE); } +#if CONFIG_MACF + error = mac_file_check_create(proc_ucred(p)); + if (error) { + if (!locked) + proc_fdunlock(p); + return (error); + } +#endif + /* * Allocate a new file descriptor. * If the process has file descriptor zero open, add to the list @@ -1883,30 +4550,41 @@ falloc_locked(p, resultfp, resultfd, locked) */ proc_fdunlock(p); - MALLOC_ZONE(fp, struct fileproc *, sizeof(struct fileproc), M_FILEPROC, M_WAITOK); + fp = (*fp_zalloc)(crarg); + if (fp == NULL) { + if (locked) + proc_fdlock(p); + return (ENOMEM); + } MALLOC_ZONE(fg, struct fileglob *, sizeof(struct fileglob), M_FILEGLOB, M_WAITOK); - bzero(fp, sizeof(struct fileproc)); + 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); fp->f_iocount = 1; fg->fg_count = 1; + fg->fg_ops = &uninitops; fp->f_fglob = fg; +#if CONFIG_MACF + mac_file_label_init(fg); +#endif - proc_fdlock(p); + kauth_cred_ref(ctx->vc_ucred); - fp->f_cred = kauth_cred_proc_ref(p); + proc_fdlock(p); - lck_mtx_lock(file_flist_lock); + fp->f_cred = ctx->vc_ucred; - nfiles++; +#if CONFIG_MACF + mac_file_label_associate(fp->f_cred, fg); +#endif - if ( (fq = p->p_fd->fd_ofiles[0]) ) { - LIST_INSERT_AFTER(fq->f_fglob, fg, f_list); - } else { - LIST_INSERT_HEAD(&filehead, fg, f_list); - } - lck_mtx_unlock(file_flist_lock); + OSAddAtomic(1, &nfiles); p->p_fd->fd_ofiles[nfd] = fp; @@ -1921,69 +4599,156 @@ falloc_locked(p, resultfp, resultfd, locked) return (0); } + /* - * Free a file structure. + * 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(fg) - struct fileglob *fg; +fg_free(struct fileglob *fg) { - kauth_cred_t cred; + OSAddAtomic(-1, &nfiles); - lck_mtx_lock(file_flist_lock); - LIST_REMOVE(fg, f_list); - nfiles--; - lck_mtx_unlock(file_flist_lock); + if (fg->fg_vn_data) { + fg_vn_data_free(fg->fg_vn_data); + fg->fg_vn_data = NULL; + } - cred = fg->fg_cred; - if (cred != NOCRED) { - fg->fg_cred = NOCRED; - kauth_cred_rele(cred); + 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); } + +/* + * fdexec + * + * Description: Perform close-on-exec processing for all files in a process + * that are either marked as close-on-exec, or which were in the + * process of being opened at the time of the execve + * + * Also handles the case (via posix_spawn()) where -all- + * files except those marked with "inherit" as treated as + * close-on-exec. + * + * Parameters: p Pointer to process calling + * execve + * + * Returns: void + * + * Locks: This function internally takes and drops proc_fdlock() + * + */ void -fdexec(p) - struct proc *p; +fdexec(proc_t p, short flags) { struct filedesc *fdp = p->p_fd; - int i = fdp->fd_lastfile; - struct fileproc **fpp = &fdp->fd_ofiles[i]; - char *flags = &fdp->fd_ofileflags[i]; - int funnel_state; + int i; + boolean_t cloexec_default = (flags & POSIX_SPAWN_CLOEXEC_DEFAULT) != 0; - funnel_state = thread_funnel_set(kernel_flock, FALSE); proc_fdlock(p); + for (i = fdp->fd_lastfile; i >= 0; i--) { - while (i >= 0) { - if ((*flags & (UF_RESERVED|UF_EXCLOSE)) == UF_EXCLOSE) { - struct fileproc *fp = *fpp; + struct fileproc *fp = fdp->fd_ofiles[i]; + char *flagp = &fdp->fd_ofileflags[i]; + if (fp && cloexec_default) { + /* + * Reverse the usual semantics of file descriptor + * inheritance - all of them should be closed + * except files marked explicitly as "inherit" and + * not marked close-on-exec. + */ + if ((*flagp & (UF_EXCLOSE|UF_INHERIT)) != UF_INHERIT) + *flagp |= UF_EXCLOSE; + *flagp &= ~UF_INHERIT; + } + + if ( + ((*flagp & (UF_RESERVED|UF_EXCLOSE)) == UF_EXCLOSE) +#if CONFIG_MACF + || (fp && mac_file_check_inherit(proc_ucred(p), fp->f_fglob)) +#endif + ) { if (i < fdp->fd_knlistsize) knote_fdclose(p, i); - - *fpp = NULL; *flags = 0; + 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); + } + closef_locked(fp, fp->f_fglob, p); - FREE_ZONE(fp, sizeof *fp, M_FILEPROC); - } - i--; fpp--; flags--; + fileproc_free(fp); + } } proc_fdunlock(p); - thread_funnel_set(kernel_flock, funnel_state); } + /* - * Copy a filedesc structure. + * fdcopy + * + * Description: Copy a filedesc structure. This is normally used as part of + * forkproc() when forking a new process, to copy the per process + * open file table over to the new process. + * + * Parameters: p Process whose open file table + * is to be copied (parent) + * uth_cdir Per thread current working + * cirectory, or NULL + * + * Returns: NULL Copy failed + * !NULL Pointer to new struct filedesc + * + * Locks: This function internally takes and drops proc_fdlock() + * + * Notes: Files are copied directly, ignoring the new resource limits + * for the process that's being copied into. Since the descriptor + * references are just additional references, this does not count + * against the number of open files on the system. + * + * The struct filedesc includes the current working directory, + * and the current root directory, if the process is chroot'ed. + * + * If the exec was called by a thread using a per thread current + * working directory, we inherit the working directory from the + * thread making the call, rather than from the process. + * + * In the case of a failure to obtain a reference, for most cases, + * the file entry will be silently dropped. There's an exception + * for the case of a chroot dir, since a failure to to obtain a + * reference there would constitute an "escape" from the chroot + * environment, which must not be allowed. In that case, we will + * deny the execve() operation, rather than allowing the escape. */ struct filedesc * -fdcopy(p) - struct proc *p; +fdcopy(proc_t p, vnode_t uth_cdir) { struct filedesc *newfdp, *fdp = p->p_fd; int i; @@ -1991,7 +4756,7 @@ fdcopy(p) vnode_t v_dir; MALLOC_ZONE(newfdp, struct filedesc *, - sizeof *newfdp, M_FILEDESC, M_WAITOK); + sizeof(*newfdp), M_FILEDESC, M_WAITOK); if (newfdp == NULL) return(NULL); @@ -2000,10 +4765,18 @@ fdcopy(p) /* * the FD_CHROOT flag will be inherited via this copy */ - (void) memcpy(newfdp, fdp, sizeof *newfdp); + (void) memcpy(newfdp, fdp, sizeof(*newfdp)); + + /* + * If we are running with per-thread current working directories, + * inherit the new current working directory from the current thread + * instead, before we take our references. + */ + if (uth_cdir != NULLVP) + 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 * a valid reference... if we can't, than set * set the pointer(s) to NULL in the child... this * will keep us from using a non-referenced vp @@ -2026,7 +4799,7 @@ fdcopy(p) * our reference from the parent also * since the vnode has gone DEAD making * it useless... by dropping it we'll - * be that much closer to recyling it + * be that much closer to recycling it */ vnode_rele(fdp->fd_cdir); fdp->fd_cdir = NULL; @@ -2037,23 +4810,25 @@ fdcopy(p) if ( (vnode_ref(v_dir)) ) newfdp->fd_rdir = NULL; vnode_put(v_dir); - } else + } else { newfdp->fd_rdir = NULL; + } } + /* Coming from a chroot environment and unable to get a reference... */ if (newfdp->fd_rdir == NULL && fdp->fd_rdir) { /* - * we couldn't get a new reference on - * the root directory being - * inherited... we might as well drop - * our reference from the parent also - * since the vnode has gone DEAD making - * it useless... by dropping it we'll - * be that much closer to recyling it + * We couldn't get a new reference on + * the chroot directory being + * inherited... this is fatal, since + * otherwise it would constitute an + * escape from a chroot environment by + * the new process. */ - vnode_rele(fdp->fd_rdir); - fdp->fd_rdir = NULL; + if (newfdp->fd_cdir) + vnode_rele(newfdp->fd_cdir); + FREE_ZONE(newfdp, sizeof *newfdp, M_FILEDESC); + return(NULL); } - newfdp->fd_refcnt = 1; /* * If the number of open files fits in the internal arrays @@ -2070,7 +4845,7 @@ fdcopy(p) * allowing the table to shrink. */ i = newfdp->fd_nfiles; - while (i > 2 * NDEXTENT && i > newfdp->fd_lastfile * 2) + while (i > 1 + 2 * NDEXTENT && i > 1 + newfdp->fd_lastfile * 2) i /= 2; } proc_fdunlock(p); @@ -2083,9 +4858,10 @@ fdcopy(p) if (newfdp->fd_rdir) vnode_rele(newfdp->fd_rdir); - FREE_ZONE(newfdp, sizeof *newfdp, M_FILEDESC); + FREE_ZONE(newfdp, sizeof(*newfdp), M_FILEDESC); return(NULL); } + (void) memset(newfdp->fd_ofiles, 0, i * OFILESIZE); proc_fdlock(p); newfdp->fd_ofileflags = (char *) &newfdp->fd_ofiles[i]; @@ -2096,18 +4872,23 @@ fdcopy(p) char *flags; (void) memcpy(newfdp->fd_ofiles, fdp->fd_ofiles, - i * sizeof *fdp->fd_ofiles); + (newfdp->fd_lastfile + 1) * sizeof(*fdp->fd_ofiles)); (void) memcpy(newfdp->fd_ofileflags, fdp->fd_ofileflags, - i * sizeof *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]; - for (i = newfdp->fd_lastfile; i >= 0; i--, fpp--) { + 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; } @@ -2122,33 +4903,57 @@ fdcopy(p) fpp = newfdp->fd_ofiles; flags = newfdp->fd_ofileflags; - for (i = newfdp->fd_lastfile; i-- >= 0; fpp++, flags++) - if ((ofp = *fpp) != NULL && !(*flags & UF_RESERVED)) { - MALLOC_ZONE(fp, struct fileproc *, sizeof(struct fileproc), M_FILEPROC, M_WAITOK); - bzero(fp, sizeof(struct fileproc)); - fp->f_flags = ofp->f_flags; - //fp->f_iocount = ofp->f_iocount; - fp->f_iocount = 0; - fp->f_fglob = ofp->f_fglob; - (void)fg_ref(fp); - *fpp = fp; + for (i = newfdp->fd_lastfile + 1; --i >= 0; fpp++, flags++) + if ((ofp = *fpp) != NULL && + 0 == (ofp->f_fglob->fg_lflags & FG_CONFINED) && + 0 == (*flags & (UF_FORKCLOSE|UF_RESERVED))) { +#if DEBUG + if (FILEPROC_TYPE(ofp) != FTYPE_SIMPLE) + panic("complex fileproc"); +#endif + fp = fileproc_alloc_init(NULL); + if (fp == NULL) { + /* + * XXX no room to copy, unable to + * XXX safely unwind state at present + */ + *fpp = NULL; + } else { + fp->f_flags |= + (ofp->f_flags & ~FP_TYPEMASK); + fp->f_fglob = ofp->f_fglob; + (void)fg_ref(fp); + *fpp = fp; + } } else { + if (i < newfdp->fd_freefile) + newfdp->fd_freefile = i; *fpp = NULL; *flags = 0; } - } else - (void) memset(newfdp->fd_ofiles, 0, i * OFILESIZE); + } proc_fdunlock(p); return (newfdp); } + /* - * Release a filedesc structure. + * fdfree + * + * Description: Release a filedesc (per process open file table) structure; + * this is done on process exit(), or from forkproc_free() if + * the fork fails for some reason subsequent to a successful + * call to fdcopy() + * + * Parameters: p Pointer to process going away + * + * Returns: void + * + * Locks: This function internally takes and drops proc_fdlock() */ void -fdfree(p) - struct proc *p; +fdfree(proc_t p) { struct filedesc *fdp; struct fileproc *fp; @@ -2156,37 +4961,31 @@ fdfree(p) proc_fdlock(p); - /* Certain daemons might not have file descriptors */ - fdp = p->p_fd; - - if ((fdp == NULL) || (--fdp->fd_refcnt > 0)) { + if (p == kernproc || NULL == (fdp = p->p_fd)) { proc_fdunlock(p); return; } - if (fdp->fd_refcnt == 0xffff) - panic("fdfree: bad fd_refcnt"); - /* Last reference: the structure can't change out from under us */ + extern struct filedesc filedesc0; + + if (&filedesc0 == fdp) + panic("filedesc0"); 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\n"); + panic("fdfree: found fp with UF_RESERVED"); - /* closef drops the iocount ... */ - if ((fp->f_flags & FP_INCHRREAD) != 0) - fp->f_iocount++; - fdp->fd_ofiles[i] = NULL; - fdp->fd_ofileflags[i] |= UF_RESERVED; + procfdtbl_reservefd(p, i); if (i < fdp->fd_knlistsize) knote_fdclose(p, i); if (fp->f_flags & FP_WAITEVENT) (void)waitevent_close(p, fp); (void) closef_locked(fp, fp->f_fglob, p); - FREE_ZONE(fp, sizeof *fp, M_FILEPROC); + fileproc_free(fp); } } FREE_ZONE(fdp->fd_ofiles, fdp->fd_nfiles * OFILESIZE, M_OFILETABL); @@ -2201,7 +5000,7 @@ fdfree(p) if (fdp->fd_rdir) vnode_rele(fdp->fd_rdir); - proc_fdlock(p); + proc_fdlock_spin(p); p->p_fd = NULL; proc_fdunlock(p); @@ -2210,72 +5009,31 @@ fdfree(p) if (fdp->fd_knhash) FREE(fdp->fd_knhash, M_KQUEUE); - FREE_ZONE(fdp, sizeof *fdp, M_FILEDESC); -} - -static int -closef_finish(fp, fg, p) - struct fileproc *fp; - struct fileglob *fg; - struct proc *p; -{ - struct vnode *vp; - struct flock lf; - int error; - struct vfs_context context; - - if ((fg->fg_flag & FHASLOCK) && fg->fg_type == DTYPE_VNODE) { - lf.l_whence = SEEK_SET; - lf.l_start = 0; - lf.l_len = 0; - lf.l_type = F_UNLCK; - vp = (struct vnode *)fg->fg_data; - context.vc_proc = p; - context.vc_ucred = fg->fg_cred; - - (void) VNOP_ADVLOCK(vp, (caddr_t)fg, F_UNLCK, &lf, F_FLOCK, &context); - } - if (fg->fg_ops) - error = fo_close(fg, p); - else - error = 0; - - if (((fp != (struct fileproc *)0) && ((fp->f_flags & FP_INCHRREAD) != 0))) { - proc_fdlock(p); - if ( ((fp->f_flags & FP_INCHRREAD) != 0) ) { - fileproc_drain(p, fp); - } - proc_fdunlock(p); - } - fg_free(fg); - - return (error); + FREE_ZONE(fdp, sizeof(*fdp), M_FILEDESC); } -int -closef(fg, p) - struct fileglob *fg; - struct proc *p; -{ - int error; - - proc_fdlock(p); - error = closef_locked((struct fileproc *)0, fg, p); - proc_fdunlock(p); - - return(error); -} /* - * Internal form of close. - * Decrement reference count on file structure. - * Note: p may be NULL when closing a file - * that was being passed in a message. + * 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(fp, fg, p) - struct fileproc *fp; - struct fileglob *fg; - struct proc *p; +closef_locked(struct fileproc *fp, struct fileglob *fg, proc_t p) { struct vnode *vp; struct flock lf; @@ -2285,6 +5043,14 @@ closef_locked(fp, fg, p) 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 @@ -2293,7 +5059,8 @@ closef_locked(fp, fg, p) * If the descriptor was in a message, POSIX-style locks * aren't passed with the descriptor. */ - if (p && (p->p_ladvflag & P_LADVLOCK) && fg->fg_type == DTYPE_VNODE) { + if (p && (p->p_ladvflag & P_LADVLOCK) && + DTYPE_VNODE == FILEGLOB_DTYPE(fg)) { proc_fdunlock(p); lf.l_whence = SEEK_SET; @@ -2303,23 +5070,22 @@ closef_locked(fp, fg, p) vp = (struct vnode *)fg->fg_data; if ( (error = vnode_getwithref(vp)) == 0 ) { - context.vc_proc = p; - context.vc_ucred = fg->fg_cred; - (void) VNOP_ADVLOCK(vp, (caddr_t)p, F_UNLCK, &lf, F_POSIX, &context); - + (void) VNOP_ADVLOCK(vp, (caddr_t)p, F_UNLCK, &lf, F_POSIX, &context, NULL); (void)vnode_put(vp); } proc_fdlock(p); } - lck_mtx_lock(&fg->fg_lock); + 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: being freed with bad fg_count (%d)", fg, fg->fg_count); + 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; @@ -2327,63 +5093,145 @@ closef_locked(fp, fg, p) fg->fg_lflags |= FG_TERM; lck_mtx_unlock(&fg->fg_lock); - proc_fdunlock(p); - error = closef_finish(fp, fg, p); - proc_fdlock(p); + if (p) + proc_fdunlock(p); + + /* 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); + + if (p) + proc_fdlock(p); return(error); } -extern int selwait; +/* + * fileproc_drain + * + * Description: Drain out pending I/O operations + * + * Parameters: p Process closing this file + * fp fileproc struct for the open + * instance on the file + * + * Returns: void + * + * Locks: Assumes the caller holds the proc_fdlock + * + * Notes: For character devices, this occurs on the last close of the + * device; for all other file descriptors, this occurs on each + * 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 + * regarding their use and interaction with this function. + */ void -fileproc_drain(struct proc *p, struct fileproc * fp) +fileproc_drain(proc_t p, struct fileproc * fp) { + struct vfs_context context; + + context.vc_thread = proc_thread(p); /* XXX */ + context.vc_ucred = fp->f_fglob->fg_cred; + fp->f_iocount-- ; /* (the one the close holds) */ while (fp->f_iocount) { - if (((fp->f_flags & FP_INSELECT)== FP_INSELECT)) { - wait_queue_wakeup_all((wait_queue_t)fp->f_waddr, &selwait, THREAD_INTERRUPTED); - } else { - if (fp->f_fglob->fg_ops->fo_drain) { - (*fp->f_fglob->fg_ops->fo_drain)(fp, p); - } + + 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); + } + if ((fp->f_flags & FP_SELCONFLICT) == FP_SELCONFLICT) { + if (waitq_wakeup64_all(&select_conflict_queue, NO_EVENT64, + THREAD_INTERRUPTED, WAITQ_ALL_PRIORITIES) == KERN_INVALID_ARGUMENT) + panic("bad select_conflict_queue"); } p->p_fpdrainwait = 1; - msleep(&p->p_fpdrainwait, &p->p_fdmlock, PRIBIO, "fpdrain",0); + msleep(&p->p_fpdrainwait, &p->p_fdmlock, PRIBIO, "fpdrain", NULL); - //panic("successful wait after drain\n"); } +#if DIAGNOSTIC + if ((fp->f_flags & FP_INSELECT) != 0) + panic("FP_INSELECT set on drained fp"); +#endif + if ((fp->f_flags & FP_SELCONFLICT) == FP_SELCONFLICT) + fp->f_flags &= ~FP_SELCONFLICT; } + +/* + * fp_free + * + * Description: Release the fd and free the fileproc associated with the fd + * in the per process open file table of the specified process; + * these values must correspond. + * + * 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 -fp_free(struct proc * p, int fd, struct fileproc * fp) +fp_free(proc_t p, int fd, struct fileproc * fp) { - proc_fdlock(p); + proc_fdlock_spin(p); fdrelse(p, fd); proc_fdunlock(p); fg_free(fp->f_fglob); - FREE_ZONE(fp, sizeof *fp, M_FILEPROC); + fileproc_free(fp); + return(0); } /* - * Apply an advisory lock on a file descriptor. + * flock * - * Just attempt to get a record lock of the requested type on - * the entire file (l_whence = SEEK_SET, l_start = 0, l_len = 0). + * Description: Apply an advisory lock on a file descriptor. + * + * Parameters: p Process making request + * uap->fd fd on which the lock is to be + * attempted + * uap->how (Un)Lock bits, including type + * retval Pointer to the call return area + * + * Returns: 0 Success + * fp_getfvp:EBADF Bad file descriptor + * fp_getfvp:ENOTSUP fd does not refer to a vnode + * vnode_getwithref:??? + * VNOP_ADVLOCK:??? + * + * Implicit returns: + * *retval (modified) Size of dtable + * + * Notes: Just attempt to get a record lock of the requested type on + * the entire file (l_whence = SEEK_SET, l_start = 0, l_len = 0). */ int -flock(struct proc *p, register struct flock_args *uap, __unused register_t *retval) +flock(proc_t p, struct flock_args *uap, __unused int32_t *retval) { int fd = uap->fd; int how = uap->how; struct fileproc *fp; struct vnode *vp; struct flock lf; - struct vfs_context context; + vfs_context_t ctx = vfs_context_current(); int error=0; AUDIT_ARG(fd, uap->fd); @@ -2395,16 +5243,13 @@ flock(struct proc *p, register struct flock_args *uap, __unused register_t *retv } AUDIT_ARG(vnpath, vp, ARG_VNODE1); - context.vc_proc = p; - context.vc_ucred = fp->f_cred; - lf.l_whence = SEEK_SET; lf.l_start = 0; 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, &context); + error = VNOP_ADVLOCK(vp, (caddr_t)fp->f_fglob, F_UNLCK, &lf, F_FLOCK, ctx, NULL); goto out; } if (how & LOCK_EX) @@ -2415,12 +5260,17 @@ flock(struct proc *p, register struct flock_args *uap, __unused register_t *retv error = EBADF; goto out; } +#if CONFIG_MACF + error = mac_file_check_lock(proc_ucred(p), fp->f_fglob, F_SETLK, &lf); + if (error) + goto out; +#endif fp->f_flag |= FHASLOCK; if (how & LOCK_NB) { - error = VNOP_ADVLOCK(vp, (caddr_t)fp->f_fglob, F_SETLK, &lf, F_FLOCK, &context); + error = VNOP_ADVLOCK(vp, (caddr_t)fp->f_fglob, F_SETLK, &lf, F_FLOCK, ctx, NULL); goto out; } - error = VNOP_ADVLOCK(vp, (caddr_t)fp->f_fglob, F_SETLK, &lf, F_FLOCK|F_WAIT, &context); + error = VNOP_ADVLOCK(vp, (caddr_t)fp->f_fglob, F_SETLK, &lf, F_FLOCK|F_WAIT, ctx, NULL); out: (void)vnode_put(vp); out1: @@ -2430,42 +5280,215 @@ out1: } /* - * File Descriptor pseudo-device driver (/dev/fd/). + * fileport_makeport * - * Opening minor device N dup()s the file (if any) connected to file - * descriptor N belonging to the calling process. Note that this driver - * consists of only the ``open()'' routine, because all subsequent - * references to this file will be direct to the other driver. + * 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. + * + * 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. + * + * Implicit returns: + * On success, name of send right is stored at user-specified address. */ int -fdopen(dev_t dev, __unused int mode, __unused int type, struct proc *p) +fileport_makeport(proc_t p, struct fileport_makeport_args *uap, + __unused int *retval) { + int err; + int fd = uap->fd; + user_addr_t user_portaddr = uap->portnamep; + struct fileproc *fp = FILEPROC_NULL; + struct fileglob *fg = NULL; + ipc_port_t fileport; + mach_port_name_t name = MACH_PORT_NULL; - /* - * XXX Kludge: set curproc->p_dupfd to contain the value of the - * the file descriptor being sought for duplication. The error - * return ensures that the vnode for this device will be released - * by vn_open. Open will detect this special error and take the - * actions in dupfdopen below. Other callers of vn_open or vnop_open - * will simply report the error. - */ - p->p_dupfd = minor(dev); - return (ENODEV); + proc_fdlock(p); + err = fp_lookup(p, fd, &fp, 1); + if (err != 0) { + goto out_unlock; + } + + if (!file_issendable(p, fp)) { + err = EINVAL; + goto out_unlock; + } + + if (FP_ISGUARDED(fp, GUARD_FILEPORT)) { + 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); + + proc_fdunlock(p); + + /* Allocate and initialize a port */ + fileport = fileport_alloc(fg); + if (fileport == IPC_PORT_NULL) { + err = EAGAIN; + fg_drop(fp); + goto out; + } + + /* Add an entry. Deallocates port on failure. */ + name = ipc_port_copyout_send(fileport, get_task_ipcspace(p->task)); + if (!MACH_PORT_VALID(name)) { + err = EINVAL; + goto out; + } + + err = copyout(&name, user_portaddr, sizeof(mach_port_name_t)); + if (err != 0) { + goto out; + } + + /* Tag the fileglob for debugging purposes */ + lck_mtx_lock_spin(&fg->fg_lock); + fg->fg_lflags |= FG_PORTMADE; + lck_mtx_unlock(&fg->fg_lock); + + fp_drop(p, fd, fp, 0); + + return 0; + +out_unlock: + proc_fdunlock(p); +out: + if (MACH_PORT_VALID(name)) { + /* Don't care if another thread races us to deallocate the entry */ + (void) mach_port_deallocate(get_task_ipcspace(p->task), name); + } + + if (fp != FILEPROC_NULL) { + fp_drop(p, fd, fp, 0); + } + + return err; +} + +void +fileport_releasefg(struct fileglob *fg) +{ + (void)closef_locked(NULL, fg, PROC_NULL); + + return; } + /* - * Duplicate the specified descriptor to a free descriptor. + * 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 -dupfdopen(fdp, indx, dfd, mode, error) - register struct filedesc *fdp; - register int indx, dfd; - int mode; - int error; +fileport_makefd(proc_t p, struct fileport_makefd_args *uap, int32_t *retval) +{ + 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; + 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; + goto out; + } + + fp = fileproc_alloc_init(NULL); + if (fp == FILEPROC_NULL) { + err = ENOMEM; + goto out; + } + + fp->f_fglob = fg; + fg_ref(fp); + + proc_fdlock(p); + err = fdalloc(p, 0, &fd); + if (err != 0) { + proc_fdunlock(p); + goto out; + } + *fdflags(p, fd) |= UF_EXCLOSE; + + procfdtbl_releasefd(p, fd, fp); + proc_fdunlock(p); + + *retval = fd; + err = 0; +out: + if ((fp != NULL) && (0 != err)) { + fileproc_free(fp); + } + + if (IPC_PORT_NULL != port) { + ipc_port_release_send(port); + } + + return err; +} + + +/* + * dupfdopen + * + * Description: Duplicate the specified descriptor to a free descriptor; + * this is the second half of fdopen(), above. + * + * Parameters: fdp filedesc pointer to fill in + * indx fd to dup to + * dfd fd to dup from + * mode mode to set on new fd + * error command code + * + * Returns: 0 Success + * EBADF Source fd is bad + * EACCES Requested mode not allowed + * !0 'error', if not ENODEV or + * ENXIO + * + * Notes: XXX This is not thread safe; see fdopen() above + */ +int +dupfdopen(struct filedesc *fdp, int indx, int dfd, int flags, int error) { struct fileproc *wfp; struct fileproc *fp; - struct proc * p = current_proc(); +#if CONFIG_MACF + int myerror; +#endif + proc_t p = current_proc(); /* * If the to-be-dup'd fd number is greater than the allowed number @@ -2484,6 +5507,13 @@ dupfdopen(fdp, indx, dfd, mode, error) proc_fdunlock(p); return (EBADF); } +#if CONFIG_MACF + myerror = mac_file_check_dup(proc_ucred(p), wfp->f_fglob, dfd); + if (myerror) { + proc_fdunlock(p); + return (myerror); + } +#endif /* * There are two cases of interest here. * @@ -2498,11 +5528,16 @@ dupfdopen(fdp, indx, dfd, mode, error) */ switch (error) { case ENODEV: + if (FP_ISGUARDED(wfp, GUARD_DUP)) { + proc_fdunlock(p); + return (EPERM); + } + /* * Check that the mode the file is being opened for is a * subset of the mode of the existing descriptor. */ - if (((mode & (FREAD|FWRITE)) | wfp->f_flag) != wfp->f_flag) { + if (((flags & (FREAD|FWRITE)) | wfp->f_flag) != wfp->f_flag) { proc_fdunlock(p); return (EACCES); } @@ -2514,29 +5549,10 @@ dupfdopen(fdp, indx, dfd, mode, error) fg_free(fp->f_fglob); fp->f_fglob = wfp->f_fglob; - fdp->fd_ofileflags[indx] = fdp->fd_ofileflags[dfd]; - - proc_fdunlock(p); - return (0); - - case ENXIO: - /* - * Steal away the file pointer from dfd, and stuff it into indx. - */ - if (indx > fdp->fd_lastfile) - fdp->fd_lastfile = indx; - - if (fp->f_fglob) - fg_free(fp->f_fglob); - fp->f_fglob = wfp->f_fglob; - - fdp->fd_ofileflags[indx] = fdp->fd_ofileflags[dfd]; - _fdrelse(fdp, dfd); + fdp->fd_ofileflags[indx] = fdp->fd_ofileflags[dfd] | + (flags & O_CLOEXEC) ? UF_EXCLOSE : 0; proc_fdunlock(p); - - FREE_ZONE(wfp, sizeof *fp, M_FILEPROC); - return (0); default: @@ -2546,6 +5562,19 @@ dupfdopen(fdp, indx, dfd, mode, error) /* 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) { @@ -2553,45 +5582,100 @@ fg_ref(struct fileproc * fp) fg = fp->f_fglob; - lck_mtx_lock(&fg->fg_lock); + 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(&fg->fg_lock); + lck_mtx_lock_spin(&fg->fg_lock); fg->fg_count--; lck_mtx_unlock(&fg->fg_lock); } - -void -fg_insertuipc(struct fileglob * fg) +#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) { -int insertque = 0; - - lck_mtx_lock(&fg->fg_lock); + 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", 0); + 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; - insertque=1; + insert = TRUE; } lck_mtx_unlock(&fg->fg_lock); + return (insert); +} - if (insertque) { - lck_mtx_lock(uipc_lock); +/* + * 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); @@ -2602,28 +5686,61 @@ int insertque = 0; } lck_mtx_unlock(&fg->fg_lock); } - } -void -fg_removeuipc(struct fileglob * fg) +/* + * 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) { -int removeque = 0; + boolean_t remove = FALSE; - lck_mtx_lock(&fg->fg_lock); + 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", 0); + msleep(&fg->fg_lflags, &fg->fg_lock, 0, "fg_removeuipc", NULL); } fg->fg_msgcount--; if (fg->fg_msgcount == 0) { fg->fg_lflags |= FG_RMMSGQ; - removeque=1; + remove = TRUE; } lck_mtx_unlock(&fg->fg_lock); + return (remove); +} - if (removeque) { - lck_mtx_lock(uipc_lock); +/* + * 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); @@ -2635,46 +5752,189 @@ int removeque = 0; lck_mtx_unlock(&fg->fg_lock); } } +#endif /* SOCKETS */ - +/* + * fo_read + * + * Description: Generic fileops read indirected through the fileops pointer + * in the fileproc structure + * + * Parameters: fp fileproc structure pointer + * uio user I/O structure pointer + * flags FOF_ flags + * ctx VFS context for operation + * + * Returns: 0 Success + * !0 Errno from read + */ int -fo_read(struct fileproc *fp, struct uio *uio, kauth_cred_t cred, int flags, struct proc *p) +fo_read(struct fileproc *fp, struct uio *uio, int flags, vfs_context_t ctx) { - return ((*fp->f_ops->fo_read)(fp, uio, cred, flags, p)); + return ((*fp->f_ops->fo_read)(fp, uio, flags, ctx)); } + +/* + * fo_write + * + * Description: Generic fileops write indirected through the fileops pointer + * in the fileproc structure + * + * Parameters: fp fileproc structure pointer + * uio user I/O structure pointer + * flags FOF_ flags + * ctx VFS context for operation + * + * Returns: 0 Success + * !0 Errno from write + */ int -fo_write(struct fileproc *fp, struct uio *uio, kauth_cred_t cred, int flags, struct proc *p) +fo_write(struct fileproc *fp, struct uio *uio, int flags, vfs_context_t ctx) { - return((*fp->f_ops->fo_write)(fp, uio, cred, flags, p)); + return((*fp->f_ops->fo_write)(fp, uio, flags, ctx)); } + +/* + * fo_ioctl + * + * Description: Generic fileops ioctl indirected through the fileops pointer + * in the fileproc structure + * + * Parameters: fp fileproc structure pointer + * com ioctl command + * data pointer to internalized copy + * of user space ioctl command + * parameter data in kernel space + * ctx VFS context for operation + * + * Returns: 0 Success + * !0 Errno from ioctl + * + * Locks: The caller is assumed to have held the proc_fdlock; this + * function releases and reacquires this lock. If the caller + * accesses data protected by this lock prior to calling this + * function, it will need to revalidate/reacquire any cached + * protected data obtained prior to the call. + */ int -fo_ioctl(struct fileproc *fp, u_long com, caddr_t data, struct proc *p) +fo_ioctl(struct fileproc *fp, u_long com, caddr_t data, vfs_context_t ctx) { -int error; + int error; - proc_fdunlock(p); - error = (*fp->f_ops->fo_ioctl)(fp, com, data, p); - proc_fdlock(p); + proc_fdunlock(vfs_context_proc(ctx)); + error = (*fp->f_ops->fo_ioctl)(fp, com, data, ctx); + proc_fdlock(vfs_context_proc(ctx)); return(error); } + +/* + * fo_select + * + * Description: Generic fileops select indirected through the fileops pointer + * in the fileproc structure + * + * Parameters: fp fileproc structure pointer + * which select which + * wql pointer to wait queue list + * ctx VFS context for operation + * + * Returns: 0 Success + * !0 Errno from select + */ int -fo_select(struct fileproc *fp, int which, void *wql, struct proc *p) +fo_select(struct fileproc *fp, int which, void *wql, vfs_context_t ctx) { - return((*fp->f_ops->fo_select)(fp, which, wql, p)); + return((*fp->f_ops->fo_select)(fp, which, wql, ctx)); } + +/* + * fo_close + * + * Description: Generic fileops close indirected through the fileops pointer + * in the fileproc structure + * + * Parameters: fp fileproc structure pointer for + * file to close + * ctx VFS context for operation + * + * Returns: 0 Success + * !0 Errno from close + */ int -fo_close(struct fileglob *fg, struct proc *p) +fo_close(struct fileglob *fg, vfs_context_t ctx) { - return((*fg->fg_ops->fo_close)(fg, p)); + return((*fg->fg_ops->fo_close)(fg, ctx)); } + +/* + * fo_kqfilter + * + * Description: Generic fileops kqueue filter indirected through the fileops + * pointer in the fileproc structure + * + * Parameters: fp fileproc structure pointer + * kn pointer to knote to filter on + * ctx VFS context for operation + * + * Returns: 0 Success + * !0 Errno from kqueue filter + */ int -fo_kqfilter(struct fileproc *fp, struct knote *kn, struct proc *p) +fo_kqfilter(struct fileproc *fp, struct knote *kn, vfs_context_t ctx) +{ + return ((*fp->f_ops->fo_kqfilter)(fp, kn, ctx)); +} + +/* + * 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) +{ + proc_fdlock_assert(p, LCK_MTX_ASSERT_OWNED); + + switch (fp->f_type) { + case DTYPE_VNODE: + case DTYPE_SOCKET: + case DTYPE_PIPE: + case DTYPE_PSXSHM: + return (0 == (fp->f_fglob->fg_lflags & FG_CONFINED)); + default: + /* DTYPE_KQUEUE, DTYPE_FSEVENTS, DTYPE_PSXSEM */ + return FALSE; + } +} + + +struct fileproc * +fileproc_alloc_init(__unused void *arg) { - return ((*fp->f_ops->fo_kqfilter)(fp, kn, p)); + struct fileproc *fp; + + MALLOC_ZONE(fp, struct fileproc *, sizeof (*fp), M_FILEPROC, M_WAITOK); + if (fp) + bzero(fp, sizeof (*fp)); + + return (fp); } +void +fileproc_free(struct fileproc *fp) +{ + switch (FILEPROC_TYPE(fp)) { + case FTYPE_SIMPLE: + FREE_ZONE(fp, sizeof (*fp), M_FILEPROC); + break; + case FTYPE_GUARDED: + guarded_fileproc_free(fp); + break; + default: + panic("%s: corrupt fp %p flags %x", __func__, fp, fp->f_flags); + } +}