freebsd-skq/sys/compat/svr4/svr4_fcntl.c

725 lines
15 KiB
C
Raw Normal View History

/*-
* Copyright (c) 1998 Mark Newton
* Copyright (c) 1994, 1997 Christos Zoulas.
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by Christos Zoulas.
* 4. The name of the author may not be used to endorse or promote products
* derived from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
* IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
2003-06-10 21:44:29 +00:00
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include <sys/param.h>
#include <sys/capability.h>
#include <sys/systm.h>
#include <sys/file.h>
#include <sys/filedesc.h>
/*#include <sys/ioctl.h>*/
#include <sys/lock.h>
#include <sys/malloc.h>
#include <sys/mount.h>
#include <sys/mutex.h>
#include <sys/namei.h>
#include <sys/priv.h>
#include <sys/proc.h>
#include <sys/stat.h>
#include <sys/syscallsubr.h>
#include <sys/unistd.h>
#include <sys/vnode.h>
#include <sys/sysproto.h>
#include <compat/svr4/svr4.h>
#include <compat/svr4/svr4_types.h>
#include <compat/svr4/svr4_signal.h>
#include <compat/svr4/svr4_proto.h>
#include <compat/svr4/svr4_util.h>
#include <compat/svr4/svr4_fcntl.h>
#include <security/mac/mac_framework.h>
2002-03-20 05:48:58 +00:00
static int svr4_to_bsd_flags(int);
static u_long svr4_to_bsd_cmd(u_long);
static int fd_revoke(struct thread *, int);
static int fd_truncate(struct thread *, int, struct flock *);
static int bsd_to_svr4_flags(int);
static void bsd_to_svr4_flock(struct flock *, struct svr4_flock *);
static void svr4_to_bsd_flock(struct svr4_flock *, struct flock *);
static void bsd_to_svr4_flock64(struct flock *, struct svr4_flock64 *);
static void svr4_to_bsd_flock64(struct svr4_flock64 *, struct flock *);
static u_long
svr4_to_bsd_cmd(cmd)
u_long cmd;
{
switch (cmd) {
case SVR4_F_DUPFD:
return F_DUPFD;
case SVR4_F_DUP2FD:
return F_DUP2FD;
case SVR4_F_GETFD:
return F_GETFD;
case SVR4_F_SETFD:
return F_SETFD;
case SVR4_F_GETFL:
return F_GETFL;
case SVR4_F_SETFL:
return F_SETFL;
case SVR4_F_GETLK:
return F_GETLK;
case SVR4_F_SETLK:
return F_SETLK;
case SVR4_F_SETLKW:
return F_SETLKW;
default:
return -1;
}
}
static int
svr4_to_bsd_flags(l)
int l;
{
int r = 0;
r |= (l & SVR4_O_RDONLY) ? O_RDONLY : 0;
r |= (l & SVR4_O_WRONLY) ? O_WRONLY : 0;
r |= (l & SVR4_O_RDWR) ? O_RDWR : 0;
r |= (l & SVR4_O_NDELAY) ? O_NONBLOCK : 0;
r |= (l & SVR4_O_APPEND) ? O_APPEND : 0;
r |= (l & SVR4_O_SYNC) ? O_FSYNC : 0;
r |= (l & SVR4_O_NONBLOCK) ? O_NONBLOCK : 0;
r |= (l & SVR4_O_PRIV) ? O_EXLOCK : 0;
r |= (l & SVR4_O_CREAT) ? O_CREAT : 0;
r |= (l & SVR4_O_TRUNC) ? O_TRUNC : 0;
r |= (l & SVR4_O_EXCL) ? O_EXCL : 0;
r |= (l & SVR4_O_NOCTTY) ? O_NOCTTY : 0;
return r;
}
static int
bsd_to_svr4_flags(l)
int l;
{
int r = 0;
r |= (l & O_RDONLY) ? SVR4_O_RDONLY : 0;
r |= (l & O_WRONLY) ? SVR4_O_WRONLY : 0;
r |= (l & O_RDWR) ? SVR4_O_RDWR : 0;
r |= (l & O_NDELAY) ? SVR4_O_NONBLOCK : 0;
r |= (l & O_APPEND) ? SVR4_O_APPEND : 0;
r |= (l & O_FSYNC) ? SVR4_O_SYNC : 0;
r |= (l & O_NONBLOCK) ? SVR4_O_NONBLOCK : 0;
r |= (l & O_EXLOCK) ? SVR4_O_PRIV : 0;
r |= (l & O_CREAT) ? SVR4_O_CREAT : 0;
r |= (l & O_TRUNC) ? SVR4_O_TRUNC : 0;
r |= (l & O_EXCL) ? SVR4_O_EXCL : 0;
r |= (l & O_NOCTTY) ? SVR4_O_NOCTTY : 0;
return r;
}
static void
bsd_to_svr4_flock(iflp, oflp)
struct flock *iflp;
struct svr4_flock *oflp;
{
switch (iflp->l_type) {
case F_RDLCK:
oflp->l_type = SVR4_F_RDLCK;
break;
case F_WRLCK:
oflp->l_type = SVR4_F_WRLCK;
break;
case F_UNLCK:
oflp->l_type = SVR4_F_UNLCK;
break;
default:
oflp->l_type = -1;
break;
}
oflp->l_whence = (short) iflp->l_whence;
oflp->l_start = (svr4_off_t) iflp->l_start;
oflp->l_len = (svr4_off_t) iflp->l_len;
oflp->l_sysid = 0;
oflp->l_pid = (svr4_pid_t) iflp->l_pid;
}
static void
svr4_to_bsd_flock(iflp, oflp)
struct svr4_flock *iflp;
struct flock *oflp;
{
switch (iflp->l_type) {
case SVR4_F_RDLCK:
oflp->l_type = F_RDLCK;
break;
case SVR4_F_WRLCK:
oflp->l_type = F_WRLCK;
break;
case SVR4_F_UNLCK:
oflp->l_type = F_UNLCK;
break;
default:
oflp->l_type = -1;
break;
}
oflp->l_whence = iflp->l_whence;
oflp->l_start = (off_t) iflp->l_start;
oflp->l_len = (off_t) iflp->l_len;
oflp->l_pid = (pid_t) iflp->l_pid;
Add the new kernel-mode NFS Lock Manager. To use it instead of the user-mode lock manager, build a kernel with the NFSLOCKD option and add '-k' to 'rpc_lockd_flags' in rc.conf. Highlights include: * Thread-safe kernel RPC client - many threads can use the same RPC client handle safely with replies being de-multiplexed at the socket upcall (typically driven directly by the NIC interrupt) and handed off to whichever thread matches the reply. For UDP sockets, many RPC clients can share the same socket. This allows the use of a single privileged UDP port number to talk to an arbitrary number of remote hosts. * Single-threaded kernel RPC server. Adding support for multi-threaded server would be relatively straightforward and would follow approximately the Solaris KPI. A single thread should be sufficient for the NLM since it should rarely block in normal operation. * Kernel mode NLM server supporting cancel requests and granted callbacks. I've tested the NLM server reasonably extensively - it passes both my own tests and the NFS Connectathon locking tests running on Solaris, Mac OS X and Ubuntu Linux. * Userland NLM client supported. While the NLM server doesn't have support for the local NFS client's locking needs, it does have to field async replies and granted callbacks from remote NLMs that the local client has contacted. We relay these replies to the userland rpc.lockd over a local domain RPC socket. * Robust deadlock detection for the local lock manager. In particular it will detect deadlocks caused by a lock request that covers more than one blocking request. As required by the NLM protocol, all deadlock detection happens synchronously - a user is guaranteed that if a lock request isn't rejected immediately, the lock will eventually be granted. The old system allowed for a 'deferred deadlock' condition where a blocked lock request could wake up and find that some other deadlock-causing lock owner had beaten them to the lock. * Since both local and remote locks are managed by the same kernel locking code, local and remote processes can safely use file locks for mutual exclusion. Local processes have no fairness advantage compared to remote processes when contending to lock a region that has just been unlocked - the local lock manager enforces a strict first-come first-served model for both local and remote lockers. Sponsored by: Isilon Systems PR: 95247 107555 115524 116679 MFC after: 2 weeks
2008-03-26 15:23:12 +00:00
oflp->l_sysid = iflp->l_sysid;
}
static void
bsd_to_svr4_flock64(iflp, oflp)
struct flock *iflp;
struct svr4_flock64 *oflp;
{
switch (iflp->l_type) {
case F_RDLCK:
oflp->l_type = SVR4_F_RDLCK;
break;
case F_WRLCK:
oflp->l_type = SVR4_F_WRLCK;
break;
case F_UNLCK:
oflp->l_type = SVR4_F_UNLCK;
break;
default:
oflp->l_type = -1;
break;
}
oflp->l_whence = (short) iflp->l_whence;
oflp->l_start = (svr4_off64_t) iflp->l_start;
oflp->l_len = (svr4_off64_t) iflp->l_len;
Add the new kernel-mode NFS Lock Manager. To use it instead of the user-mode lock manager, build a kernel with the NFSLOCKD option and add '-k' to 'rpc_lockd_flags' in rc.conf. Highlights include: * Thread-safe kernel RPC client - many threads can use the same RPC client handle safely with replies being de-multiplexed at the socket upcall (typically driven directly by the NIC interrupt) and handed off to whichever thread matches the reply. For UDP sockets, many RPC clients can share the same socket. This allows the use of a single privileged UDP port number to talk to an arbitrary number of remote hosts. * Single-threaded kernel RPC server. Adding support for multi-threaded server would be relatively straightforward and would follow approximately the Solaris KPI. A single thread should be sufficient for the NLM since it should rarely block in normal operation. * Kernel mode NLM server supporting cancel requests and granted callbacks. I've tested the NLM server reasonably extensively - it passes both my own tests and the NFS Connectathon locking tests running on Solaris, Mac OS X and Ubuntu Linux. * Userland NLM client supported. While the NLM server doesn't have support for the local NFS client's locking needs, it does have to field async replies and granted callbacks from remote NLMs that the local client has contacted. We relay these replies to the userland rpc.lockd over a local domain RPC socket. * Robust deadlock detection for the local lock manager. In particular it will detect deadlocks caused by a lock request that covers more than one blocking request. As required by the NLM protocol, all deadlock detection happens synchronously - a user is guaranteed that if a lock request isn't rejected immediately, the lock will eventually be granted. The old system allowed for a 'deferred deadlock' condition where a blocked lock request could wake up and find that some other deadlock-causing lock owner had beaten them to the lock. * Since both local and remote locks are managed by the same kernel locking code, local and remote processes can safely use file locks for mutual exclusion. Local processes have no fairness advantage compared to remote processes when contending to lock a region that has just been unlocked - the local lock manager enforces a strict first-come first-served model for both local and remote lockers. Sponsored by: Isilon Systems PR: 95247 107555 115524 116679 MFC after: 2 weeks
2008-03-26 15:23:12 +00:00
oflp->l_sysid = iflp->l_sysid;
oflp->l_pid = (svr4_pid_t) iflp->l_pid;
}
static void
svr4_to_bsd_flock64(iflp, oflp)
struct svr4_flock64 *iflp;
struct flock *oflp;
{
switch (iflp->l_type) {
case SVR4_F_RDLCK:
oflp->l_type = F_RDLCK;
break;
case SVR4_F_WRLCK:
oflp->l_type = F_WRLCK;
break;
case SVR4_F_UNLCK:
oflp->l_type = F_UNLCK;
break;
default:
oflp->l_type = -1;
break;
}
oflp->l_whence = iflp->l_whence;
oflp->l_start = (off_t) iflp->l_start;
oflp->l_len = (off_t) iflp->l_len;
oflp->l_pid = (pid_t) iflp->l_pid;
}
static int
fd_revoke(td, fd)
struct thread *td;
int fd;
{
struct vnode *vp;
struct mount *mp;
struct vattr vattr;
int error, *retval;
retval = td->td_retval;
/*
* If we ever want to support Capsicum on SVR4 processes (unlikely)
* or FreeBSD grows a native frevoke() (more likely), we will need a
Merge Capsicum overhaul: - Capability is no longer separate descriptor type. Now every descriptor has set of its own capability rights. - The cap_new(2) system call is left, but it is no longer documented and should not be used in new code. - The new syscall cap_rights_limit(2) should be used instead of cap_new(2), which limits capability rights of the given descriptor without creating a new one. - The cap_getrights(2) syscall is renamed to cap_rights_get(2). - If CAP_IOCTL capability right is present we can further reduce allowed ioctls list with the new cap_ioctls_limit(2) syscall. List of allowed ioctls can be retrived with cap_ioctls_get(2) syscall. - If CAP_FCNTL capability right is present we can further reduce fcntls that can be used with the new cap_fcntls_limit(2) syscall and retrive them with cap_fcntls_get(2). - To support ioctl and fcntl white-listing the filedesc structure was heavly modified. - The audit subsystem, kdump and procstat tools were updated to recognize new syscalls. - Capability rights were revised and eventhough I tried hard to provide backward API and ABI compatibility there are some incompatible changes that are described in detail below: CAP_CREATE old behaviour: - Allow for openat(2)+O_CREAT. - Allow for linkat(2). - Allow for symlinkat(2). CAP_CREATE new behaviour: - Allow for openat(2)+O_CREAT. Added CAP_LINKAT: - Allow for linkat(2). ABI: Reuses CAP_RMDIR bit. - Allow to be target for renameat(2). Added CAP_SYMLINKAT: - Allow for symlinkat(2). Removed CAP_DELETE. Old behaviour: - Allow for unlinkat(2) when removing non-directory object. - Allow to be source for renameat(2). Removed CAP_RMDIR. Old behaviour: - Allow for unlinkat(2) when removing directory. Added CAP_RENAMEAT: - Required for source directory for the renameat(2) syscall. Added CAP_UNLINKAT (effectively it replaces CAP_DELETE and CAP_RMDIR): - Allow for unlinkat(2) on any object. - Required if target of renameat(2) exists and will be removed by this call. Removed CAP_MAPEXEC. CAP_MMAP old behaviour: - Allow for mmap(2) with any combination of PROT_NONE, PROT_READ and PROT_WRITE. CAP_MMAP new behaviour: - Allow for mmap(2)+PROT_NONE. Added CAP_MMAP_R: - Allow for mmap(PROT_READ). Added CAP_MMAP_W: - Allow for mmap(PROT_WRITE). Added CAP_MMAP_X: - Allow for mmap(PROT_EXEC). Added CAP_MMAP_RW: - Allow for mmap(PROT_READ | PROT_WRITE). Added CAP_MMAP_RX: - Allow for mmap(PROT_READ | PROT_EXEC). Added CAP_MMAP_WX: - Allow for mmap(PROT_WRITE | PROT_EXEC). Added CAP_MMAP_RWX: - Allow for mmap(PROT_READ | PROT_WRITE | PROT_EXEC). Renamed CAP_MKDIR to CAP_MKDIRAT. Renamed CAP_MKFIFO to CAP_MKFIFOAT. Renamed CAP_MKNODE to CAP_MKNODEAT. CAP_READ old behaviour: - Allow pread(2). - Disallow read(2), readv(2) (if there is no CAP_SEEK). CAP_READ new behaviour: - Allow read(2), readv(2). - Disallow pread(2) (CAP_SEEK was also required). CAP_WRITE old behaviour: - Allow pwrite(2). - Disallow write(2), writev(2) (if there is no CAP_SEEK). CAP_WRITE new behaviour: - Allow write(2), writev(2). - Disallow pwrite(2) (CAP_SEEK was also required). Added convinient defines: #define CAP_PREAD (CAP_SEEK | CAP_READ) #define CAP_PWRITE (CAP_SEEK | CAP_WRITE) #define CAP_MMAP_R (CAP_MMAP | CAP_SEEK | CAP_READ) #define CAP_MMAP_W (CAP_MMAP | CAP_SEEK | CAP_WRITE) #define CAP_MMAP_X (CAP_MMAP | CAP_SEEK | 0x0000000000000008ULL) #define CAP_MMAP_RW (CAP_MMAP_R | CAP_MMAP_W) #define CAP_MMAP_RX (CAP_MMAP_R | CAP_MMAP_X) #define CAP_MMAP_WX (CAP_MMAP_W | CAP_MMAP_X) #define CAP_MMAP_RWX (CAP_MMAP_R | CAP_MMAP_W | CAP_MMAP_X) #define CAP_RECV CAP_READ #define CAP_SEND CAP_WRITE #define CAP_SOCK_CLIENT \ (CAP_CONNECT | CAP_GETPEERNAME | CAP_GETSOCKNAME | CAP_GETSOCKOPT | \ CAP_PEELOFF | CAP_RECV | CAP_SEND | CAP_SETSOCKOPT | CAP_SHUTDOWN) #define CAP_SOCK_SERVER \ (CAP_ACCEPT | CAP_BIND | CAP_GETPEERNAME | CAP_GETSOCKNAME | \ CAP_GETSOCKOPT | CAP_LISTEN | CAP_PEELOFF | CAP_RECV | CAP_SEND | \ CAP_SETSOCKOPT | CAP_SHUTDOWN) Added defines for backward API compatibility: #define CAP_MAPEXEC CAP_MMAP_X #define CAP_DELETE CAP_UNLINKAT #define CAP_MKDIR CAP_MKDIRAT #define CAP_RMDIR CAP_UNLINKAT #define CAP_MKFIFO CAP_MKFIFOAT #define CAP_MKNOD CAP_MKNODAT #define CAP_SOCK_ALL (CAP_SOCK_CLIENT | CAP_SOCK_SERVER) Sponsored by: The FreeBSD Foundation Reviewed by: Christoph Mallon <christoph.mallon@gmx.de> Many aspects discussed with: rwatson, benl, jonathan ABI compatibility discussed with: kib
2013-03-02 00:53:12 +00:00
* CAP_FREVOKE here.
*
Merge Capsicum overhaul: - Capability is no longer separate descriptor type. Now every descriptor has set of its own capability rights. - The cap_new(2) system call is left, but it is no longer documented and should not be used in new code. - The new syscall cap_rights_limit(2) should be used instead of cap_new(2), which limits capability rights of the given descriptor without creating a new one. - The cap_getrights(2) syscall is renamed to cap_rights_get(2). - If CAP_IOCTL capability right is present we can further reduce allowed ioctls list with the new cap_ioctls_limit(2) syscall. List of allowed ioctls can be retrived with cap_ioctls_get(2) syscall. - If CAP_FCNTL capability right is present we can further reduce fcntls that can be used with the new cap_fcntls_limit(2) syscall and retrive them with cap_fcntls_get(2). - To support ioctl and fcntl white-listing the filedesc structure was heavly modified. - The audit subsystem, kdump and procstat tools were updated to recognize new syscalls. - Capability rights were revised and eventhough I tried hard to provide backward API and ABI compatibility there are some incompatible changes that are described in detail below: CAP_CREATE old behaviour: - Allow for openat(2)+O_CREAT. - Allow for linkat(2). - Allow for symlinkat(2). CAP_CREATE new behaviour: - Allow for openat(2)+O_CREAT. Added CAP_LINKAT: - Allow for linkat(2). ABI: Reuses CAP_RMDIR bit. - Allow to be target for renameat(2). Added CAP_SYMLINKAT: - Allow for symlinkat(2). Removed CAP_DELETE. Old behaviour: - Allow for unlinkat(2) when removing non-directory object. - Allow to be source for renameat(2). Removed CAP_RMDIR. Old behaviour: - Allow for unlinkat(2) when removing directory. Added CAP_RENAMEAT: - Required for source directory for the renameat(2) syscall. Added CAP_UNLINKAT (effectively it replaces CAP_DELETE and CAP_RMDIR): - Allow for unlinkat(2) on any object. - Required if target of renameat(2) exists and will be removed by this call. Removed CAP_MAPEXEC. CAP_MMAP old behaviour: - Allow for mmap(2) with any combination of PROT_NONE, PROT_READ and PROT_WRITE. CAP_MMAP new behaviour: - Allow for mmap(2)+PROT_NONE. Added CAP_MMAP_R: - Allow for mmap(PROT_READ). Added CAP_MMAP_W: - Allow for mmap(PROT_WRITE). Added CAP_MMAP_X: - Allow for mmap(PROT_EXEC). Added CAP_MMAP_RW: - Allow for mmap(PROT_READ | PROT_WRITE). Added CAP_MMAP_RX: - Allow for mmap(PROT_READ | PROT_EXEC). Added CAP_MMAP_WX: - Allow for mmap(PROT_WRITE | PROT_EXEC). Added CAP_MMAP_RWX: - Allow for mmap(PROT_READ | PROT_WRITE | PROT_EXEC). Renamed CAP_MKDIR to CAP_MKDIRAT. Renamed CAP_MKFIFO to CAP_MKFIFOAT. Renamed CAP_MKNODE to CAP_MKNODEAT. CAP_READ old behaviour: - Allow pread(2). - Disallow read(2), readv(2) (if there is no CAP_SEEK). CAP_READ new behaviour: - Allow read(2), readv(2). - Disallow pread(2) (CAP_SEEK was also required). CAP_WRITE old behaviour: - Allow pwrite(2). - Disallow write(2), writev(2) (if there is no CAP_SEEK). CAP_WRITE new behaviour: - Allow write(2), writev(2). - Disallow pwrite(2) (CAP_SEEK was also required). Added convinient defines: #define CAP_PREAD (CAP_SEEK | CAP_READ) #define CAP_PWRITE (CAP_SEEK | CAP_WRITE) #define CAP_MMAP_R (CAP_MMAP | CAP_SEEK | CAP_READ) #define CAP_MMAP_W (CAP_MMAP | CAP_SEEK | CAP_WRITE) #define CAP_MMAP_X (CAP_MMAP | CAP_SEEK | 0x0000000000000008ULL) #define CAP_MMAP_RW (CAP_MMAP_R | CAP_MMAP_W) #define CAP_MMAP_RX (CAP_MMAP_R | CAP_MMAP_X) #define CAP_MMAP_WX (CAP_MMAP_W | CAP_MMAP_X) #define CAP_MMAP_RWX (CAP_MMAP_R | CAP_MMAP_W | CAP_MMAP_X) #define CAP_RECV CAP_READ #define CAP_SEND CAP_WRITE #define CAP_SOCK_CLIENT \ (CAP_CONNECT | CAP_GETPEERNAME | CAP_GETSOCKNAME | CAP_GETSOCKOPT | \ CAP_PEELOFF | CAP_RECV | CAP_SEND | CAP_SETSOCKOPT | CAP_SHUTDOWN) #define CAP_SOCK_SERVER \ (CAP_ACCEPT | CAP_BIND | CAP_GETPEERNAME | CAP_GETSOCKNAME | \ CAP_GETSOCKOPT | CAP_LISTEN | CAP_PEELOFF | CAP_RECV | CAP_SEND | \ CAP_SETSOCKOPT | CAP_SHUTDOWN) Added defines for backward API compatibility: #define CAP_MAPEXEC CAP_MMAP_X #define CAP_DELETE CAP_UNLINKAT #define CAP_MKDIR CAP_MKDIRAT #define CAP_RMDIR CAP_UNLINKAT #define CAP_MKFIFO CAP_MKFIFOAT #define CAP_MKNOD CAP_MKNODAT #define CAP_SOCK_ALL (CAP_SOCK_CLIENT | CAP_SOCK_SERVER) Sponsored by: The FreeBSD Foundation Reviewed by: Christoph Mallon <christoph.mallon@gmx.de> Many aspects discussed with: rwatson, benl, jonathan ABI compatibility discussed with: kib
2013-03-02 00:53:12 +00:00
* In the meantime, use CAP_ALL: if a SVR4 process wants to
* do an frevoke(), it needs to do it on either a regular file
* descriptor or a fully-privileged capability (which is effectively
* the same as a non-capability-restricted file descriptor).
*/
Merge Capsicum overhaul: - Capability is no longer separate descriptor type. Now every descriptor has set of its own capability rights. - The cap_new(2) system call is left, but it is no longer documented and should not be used in new code. - The new syscall cap_rights_limit(2) should be used instead of cap_new(2), which limits capability rights of the given descriptor without creating a new one. - The cap_getrights(2) syscall is renamed to cap_rights_get(2). - If CAP_IOCTL capability right is present we can further reduce allowed ioctls list with the new cap_ioctls_limit(2) syscall. List of allowed ioctls can be retrived with cap_ioctls_get(2) syscall. - If CAP_FCNTL capability right is present we can further reduce fcntls that can be used with the new cap_fcntls_limit(2) syscall and retrive them with cap_fcntls_get(2). - To support ioctl and fcntl white-listing the filedesc structure was heavly modified. - The audit subsystem, kdump and procstat tools were updated to recognize new syscalls. - Capability rights were revised and eventhough I tried hard to provide backward API and ABI compatibility there are some incompatible changes that are described in detail below: CAP_CREATE old behaviour: - Allow for openat(2)+O_CREAT. - Allow for linkat(2). - Allow for symlinkat(2). CAP_CREATE new behaviour: - Allow for openat(2)+O_CREAT. Added CAP_LINKAT: - Allow for linkat(2). ABI: Reuses CAP_RMDIR bit. - Allow to be target for renameat(2). Added CAP_SYMLINKAT: - Allow for symlinkat(2). Removed CAP_DELETE. Old behaviour: - Allow for unlinkat(2) when removing non-directory object. - Allow to be source for renameat(2). Removed CAP_RMDIR. Old behaviour: - Allow for unlinkat(2) when removing directory. Added CAP_RENAMEAT: - Required for source directory for the renameat(2) syscall. Added CAP_UNLINKAT (effectively it replaces CAP_DELETE and CAP_RMDIR): - Allow for unlinkat(2) on any object. - Required if target of renameat(2) exists and will be removed by this call. Removed CAP_MAPEXEC. CAP_MMAP old behaviour: - Allow for mmap(2) with any combination of PROT_NONE, PROT_READ and PROT_WRITE. CAP_MMAP new behaviour: - Allow for mmap(2)+PROT_NONE. Added CAP_MMAP_R: - Allow for mmap(PROT_READ). Added CAP_MMAP_W: - Allow for mmap(PROT_WRITE). Added CAP_MMAP_X: - Allow for mmap(PROT_EXEC). Added CAP_MMAP_RW: - Allow for mmap(PROT_READ | PROT_WRITE). Added CAP_MMAP_RX: - Allow for mmap(PROT_READ | PROT_EXEC). Added CAP_MMAP_WX: - Allow for mmap(PROT_WRITE | PROT_EXEC). Added CAP_MMAP_RWX: - Allow for mmap(PROT_READ | PROT_WRITE | PROT_EXEC). Renamed CAP_MKDIR to CAP_MKDIRAT. Renamed CAP_MKFIFO to CAP_MKFIFOAT. Renamed CAP_MKNODE to CAP_MKNODEAT. CAP_READ old behaviour: - Allow pread(2). - Disallow read(2), readv(2) (if there is no CAP_SEEK). CAP_READ new behaviour: - Allow read(2), readv(2). - Disallow pread(2) (CAP_SEEK was also required). CAP_WRITE old behaviour: - Allow pwrite(2). - Disallow write(2), writev(2) (if there is no CAP_SEEK). CAP_WRITE new behaviour: - Allow write(2), writev(2). - Disallow pwrite(2) (CAP_SEEK was also required). Added convinient defines: #define CAP_PREAD (CAP_SEEK | CAP_READ) #define CAP_PWRITE (CAP_SEEK | CAP_WRITE) #define CAP_MMAP_R (CAP_MMAP | CAP_SEEK | CAP_READ) #define CAP_MMAP_W (CAP_MMAP | CAP_SEEK | CAP_WRITE) #define CAP_MMAP_X (CAP_MMAP | CAP_SEEK | 0x0000000000000008ULL) #define CAP_MMAP_RW (CAP_MMAP_R | CAP_MMAP_W) #define CAP_MMAP_RX (CAP_MMAP_R | CAP_MMAP_X) #define CAP_MMAP_WX (CAP_MMAP_W | CAP_MMAP_X) #define CAP_MMAP_RWX (CAP_MMAP_R | CAP_MMAP_W | CAP_MMAP_X) #define CAP_RECV CAP_READ #define CAP_SEND CAP_WRITE #define CAP_SOCK_CLIENT \ (CAP_CONNECT | CAP_GETPEERNAME | CAP_GETSOCKNAME | CAP_GETSOCKOPT | \ CAP_PEELOFF | CAP_RECV | CAP_SEND | CAP_SETSOCKOPT | CAP_SHUTDOWN) #define CAP_SOCK_SERVER \ (CAP_ACCEPT | CAP_BIND | CAP_GETPEERNAME | CAP_GETSOCKNAME | \ CAP_GETSOCKOPT | CAP_LISTEN | CAP_PEELOFF | CAP_RECV | CAP_SEND | \ CAP_SETSOCKOPT | CAP_SHUTDOWN) Added defines for backward API compatibility: #define CAP_MAPEXEC CAP_MMAP_X #define CAP_DELETE CAP_UNLINKAT #define CAP_MKDIR CAP_MKDIRAT #define CAP_RMDIR CAP_UNLINKAT #define CAP_MKFIFO CAP_MKFIFOAT #define CAP_MKNOD CAP_MKNODAT #define CAP_SOCK_ALL (CAP_SOCK_CLIENT | CAP_SOCK_SERVER) Sponsored by: The FreeBSD Foundation Reviewed by: Christoph Mallon <christoph.mallon@gmx.de> Many aspects discussed with: rwatson, benl, jonathan ABI compatibility discussed with: kib
2013-03-02 00:53:12 +00:00
if ((error = fgetvp(td, fd, CAP_ALL, &vp)) != 0)
return (error);
if (vp->v_type != VCHR && vp->v_type != VBLK) {
error = EINVAL;
goto out;
}
#ifdef MAC
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY);
error = mac_vnode_check_revoke(td->td_ucred, vp);
VOP_UNLOCK(vp, 0);
if (error)
goto out;
#endif
if ((error = VOP_GETATTR(vp, &vattr, td->td_ucred)) != 0)
goto out;
if (td->td_ucred->cr_uid != vattr.va_uid &&
(error = priv_check(td, PRIV_VFS_ADMIN)) != 0)
goto out;
if ((error = vn_start_write(vp, &mp, V_WAIT | PCATCH)) != 0)
goto out;
if (vcount(vp) > 1)
VOP_REVOKE(vp, REVOKEALL);
vn_finished_write(mp);
out:
vrele(vp);
return error;
}
static int
fd_truncate(td, fd, flp)
struct thread *td;
int fd;
struct flock *flp;
{
off_t start, length;
struct file *fp;
struct vnode *vp;
struct vattr vattr;
int error, *retval;
struct ftruncate_args ft;
retval = td->td_retval;
/*
* We only support truncating the file.
*/
if ((error = fget(td, fd, CAP_FTRUNCATE, &fp)) != 0)
return (error);
vp = fp->f_vnode;
if (fp->f_type != DTYPE_VNODE || vp->v_type == VFIFO) {
fdrop(fp, td);
return ESPIPE;
}
if ((error = VOP_GETATTR(vp, &vattr, td->td_ucred)) != 0) {
fdrop(fp, td);
return error;
}
length = vattr.va_size;
switch (flp->l_whence) {
case SEEK_CUR:
start = fp->f_offset + flp->l_start;
break;
case SEEK_END:
start = flp->l_start + length;
break;
case SEEK_SET:
start = flp->l_start;
break;
default:
fdrop(fp, td);
return EINVAL;
}
if (start + flp->l_len < length) {
/* We don't support free'ing in the middle of the file */
fdrop(fp, td);
return EINVAL;
}
2002-12-14 01:56:26 +00:00
ft.fd = fd;
ft.length = start;
error = sys_ftruncate(td, &ft);
fdrop(fp, td);
return (error);
}
int
svr4_sys_open(td, uap)
struct thread *td;
struct svr4_sys_open_args *uap;
{
struct proc *p = td->td_proc;
char *newpath;
int bsd_flags, error, retval;
CHECKALTEXIST(td, uap->path, &newpath);
bsd_flags = svr4_to_bsd_flags(uap->flags);
error = kern_open(td, newpath, UIO_SYSSPACE, bsd_flags, uap->mode);
free(newpath, M_TEMP);
if (error) {
/* uprintf("svr4_open(%s, 0x%0x, 0%o): %d\n", uap->path,
uap->flags, uap->mode, error);*/
return error;
}
retval = td->td_retval[0];
2001-01-23 21:33:55 +00:00
PROC_LOCK(p);
if (!(bsd_flags & O_NOCTTY) && SESS_LEADER(p) &&
!(p->p_flag & P_CONTROLT)) {
#if defined(NOTYET)
struct file *fp;
error = fget(td, retval, CAP_IOCTL, &fp);
2001-01-23 21:33:55 +00:00
PROC_UNLOCK(p);
/*
* we may have lost a race the above open() and
* another thread issuing a close()
*/
if (error)
return (EBADF); /* XXX: correct errno? */
/* ignore any error, just give it a try */
if (fp->f_type == DTYPE_VNODE)
fo_ioctl(fp, TIOCSCTTY, (caddr_t) 0, td->td_ucred,
td);
fdrop(fp, td);
} else {
2001-01-23 21:33:55 +00:00
PROC_UNLOCK(p);
}
2001-01-23 21:33:55 +00:00
#else
}
2001-01-23 21:33:55 +00:00
PROC_UNLOCK(p);
#endif
return error;
}
int
svr4_sys_open64(td, uap)
struct thread *td;
struct svr4_sys_open64_args *uap;
{
return svr4_sys_open(td, (struct svr4_sys_open_args *)uap);
}
int
svr4_sys_creat(td, uap)
struct thread *td;
struct svr4_sys_creat_args *uap;
{
char *newpath;
int error;
CHECKALTEXIST(td, uap->path, &newpath);
error = kern_open(td, newpath, UIO_SYSSPACE, O_WRONLY | O_CREAT |
O_TRUNC, uap->mode);
free(newpath, M_TEMP);
return (error);
}
int
svr4_sys_creat64(td, uap)
struct thread *td;
struct svr4_sys_creat64_args *uap;
{
return svr4_sys_creat(td, (struct svr4_sys_creat_args *)uap);
}
int
svr4_sys_llseek(td, uap)
struct thread *td;
2001-01-23 21:33:55 +00:00
struct svr4_sys_llseek_args *uap;
{
struct lseek_args ap;
2002-12-14 01:56:26 +00:00
ap.fd = uap->fd;
#if BYTE_ORDER == BIG_ENDIAN
2002-12-14 01:56:26 +00:00
ap.offset = (((u_int64_t) uap->offset1) << 32) |
uap->offset2;
#else
2002-12-14 01:56:26 +00:00
ap.offset = (((u_int64_t) uap->offset2) << 32) |
uap->offset1;
#endif
2002-12-14 01:56:26 +00:00
ap.whence = uap->whence;
return sys_lseek(td, &ap);
}
int
svr4_sys_access(td, uap)
struct thread *td;
struct svr4_sys_access_args *uap;
{
char *newpath;
int error;
CHECKALTEXIST(td, uap->path, &newpath);
error = kern_access(td, newpath, UIO_SYSSPACE, uap->amode);
free(newpath, M_TEMP);
return (error);
}
#if defined(NOTYET)
int
svr4_sys_pread(td, uap)
struct thread *td;
struct svr4_sys_pread_args *uap;
{
struct pread_args pra;
/*
* Just translate the args structure and call the NetBSD
* pread(2) system call (offset type is 64-bit in NetBSD).
*/
2002-12-14 01:56:26 +00:00
pra.fd = uap->fd;
pra.buf = uap->buf;
pra.nbyte = uap->nbyte;
pra.offset = uap->off;
return pread(td, &pra);
}
#endif
#if defined(NOTYET)
int
svr4_sys_pread64(td, v, retval)
struct thread *td;
void *v;
register_t *retval;
{
struct svr4_sys_pread64_args *uap = v;
struct sys_pread_args pra;
/*
* Just translate the args structure and call the NetBSD
* pread(2) system call (offset type is 64-bit in NetBSD).
*/
2002-12-14 01:56:26 +00:00
pra.fd = uap->fd;
pra.buf = uap->buf;
pra.nbyte = uap->nbyte;
pra.offset = uap->off;
return (sys_pread(td, &pra, retval));
}
#endif /* NOTYET */
#if defined(NOTYET)
int
svr4_sys_pwrite(td, uap)
struct thread *td;
struct svr4_sys_pwrite_args *uap;
{
struct pwrite_args pwa;
/*
* Just translate the args structure and call the NetBSD
* pwrite(2) system call (offset type is 64-bit in NetBSD).
*/
2002-12-14 01:56:26 +00:00
pwa.fd = uap->fd;
pwa.buf = uap->buf;
pwa.nbyte = uap->nbyte;
pwa.offset = uap->off;
return pwrite(td, &pwa);
}
#endif
#if defined(NOTYET)
int
svr4_sys_pwrite64(td, v, retval)
struct thread *td;
void *v;
register_t *retval;
{
struct svr4_sys_pwrite64_args *uap = v;
struct sys_pwrite_args pwa;
/*
* Just translate the args structure and call the NetBSD
* pwrite(2) system call (offset type is 64-bit in NetBSD).
*/
2002-12-14 01:56:26 +00:00
pwa.fd = uap->fd;
pwa.buf = uap->buf;
pwa.nbyte = uap->nbyte;
pwa.offset = uap->off;
return (sys_pwrite(td, &pwa, retval));
}
#endif /* NOTYET */
int
svr4_sys_fcntl(td, uap)
struct thread *td;
struct svr4_sys_fcntl_args *uap;
{
int cmd, error, *retval;
retval = td->td_retval;
cmd = svr4_to_bsd_cmd(uap->cmd);
switch (cmd) {
case F_DUPFD:
case F_DUP2FD:
case F_GETFD:
case F_SETFD:
return (kern_fcntl(td, uap->fd, cmd, (intptr_t)uap->arg));
case F_GETFL:
error = kern_fcntl(td, uap->fd, cmd, (intptr_t)uap->arg);
if (error)
return (error);
*retval = bsd_to_svr4_flags(*retval);
return (error);
case F_SETFL:
{
/*
* we must save the O_ASYNC flag, as that is
* handled by ioctl(_, I_SETSIG, _) emulation.
*/
int flags;
2002-12-14 01:56:26 +00:00
DPRINTF(("Setting flags %p\n", uap->arg));
error = kern_fcntl(td, uap->fd, F_GETFL, 0);
if (error)
return (error);
flags = *retval;
flags &= O_ASYNC;
2002-12-14 01:56:26 +00:00
flags |= svr4_to_bsd_flags((u_long) uap->arg);
return (kern_fcntl(td, uap->fd, F_SETFL, flags));
}
case F_GETLK:
case F_SETLK:
case F_SETLKW:
{
struct svr4_flock ifl;
struct flock fl;
error = copyin(uap->arg, &ifl, sizeof (ifl));
if (error)
return (error);
svr4_to_bsd_flock(&ifl, &fl);
error = kern_fcntl(td, uap->fd, cmd, (intptr_t)&fl);
if (error || cmd != F_GETLK)
return (error);
bsd_to_svr4_flock(&fl, &ifl);
return (copyout(&ifl, uap->arg, sizeof (ifl)));
}
case -1:
2002-12-14 01:56:26 +00:00
switch (uap->cmd) {
case SVR4_F_FREESP:
{
struct svr4_flock ifl;
struct flock fl;
2002-12-14 01:56:26 +00:00
error = copyin(uap->arg, &ifl,
sizeof ifl);
if (error)
return error;
svr4_to_bsd_flock(&ifl, &fl);
2002-12-14 01:56:26 +00:00
return fd_truncate(td, uap->fd, &fl);
}
case SVR4_F_GETLK64:
case SVR4_F_SETLK64:
case SVR4_F_SETLKW64:
{
struct svr4_flock64 ifl;
struct flock fl;
switch (uap->cmd) {
case SVR4_F_GETLK64:
cmd = F_GETLK;
break;
case SVR4_F_SETLK64:
cmd = F_SETLK;
break;
case SVR4_F_SETLKW64:
cmd = F_SETLKW;
break;
}
2002-12-14 01:56:26 +00:00
error = copyin(uap->arg, &ifl,
sizeof (ifl));
if (error)
return (error);
svr4_to_bsd_flock64(&ifl, &fl);
error = kern_fcntl(td, uap->fd, cmd,
(intptr_t)&fl);
if (error || cmd != F_GETLK)
return (error);
bsd_to_svr4_flock64(&fl, &ifl);
return (copyout(&ifl, uap->arg,
sizeof (ifl)));
}
case SVR4_F_FREESP64:
{
struct svr4_flock64 ifl;
struct flock fl;
2002-12-14 01:56:26 +00:00
error = copyin(uap->arg, &ifl,
sizeof ifl);
if (error)
return error;
svr4_to_bsd_flock64(&ifl, &fl);
2002-12-14 01:56:26 +00:00
return fd_truncate(td, uap->fd, &fl);
}
case SVR4_F_REVOKE:
2002-12-14 01:56:26 +00:00
return fd_revoke(td, uap->fd);
default:
return ENOSYS;
}
default:
return ENOSYS;
}
}