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

6 files changed, 29 insertions, 19 deletions

diff --git a/sys/compat/freebsd32/syscalls.master b/sys/compat/freebsd32/syscalls.master
index 4106447..6552d13 100644
--- a/sys/compat/freebsd32/syscalls.master
+++ b/sys/compat/freebsd32/syscalls.master
@@ -963,7 +963,7 @@
 				    struct shmid_ds32 *buf); }
 513	AUE_LPATHCONF	NOPROTO	{ int lpathconf(char *path, int name); }
 514	AUE_CAP_NEW	NOPROTO	{ int cap_new(int fd, uint64_t rights); }
-515	AUE_CAP_GETRIGHTS	NOPROTO	{ int cap_getrights(int fd, \
+515	AUE_CAP_RIGHTS_GET	NOPROTO	{ int cap_rights_get(int fd, \
 				    uint64_t *rightsp); }
 516	AUE_CAP_ENTER	NOPROTO	{ int cap_enter(void); }
 517	AUE_CAP_GETMODE	NOPROTO	{ int cap_getmode(u_int *modep); }
@@ -1005,3 +1005,13 @@
 				    struct wrusage32 *wrusage, \
 				    siginfo_t *info); }
 
+533	AUE_CAP_RIGHTS_LIMIT	NOPROTO	{ int cap_rights_limit(int fd, \
+				    uint64_t rights); }
+534	AUE_CAP_IOCTLS_LIMIT	NOPROTO	{ int cap_ioctls_limit(int fd, \
+				    const u_long *cmds, size_t ncmds); }
+535	AUE_CAP_IOCTLS_GET	NOPROTO	{ ssize_t cap_ioctls_get(int fd, \
+				    u_long *cmds, size_t maxcmds); }
+536	AUE_CAP_FCNTLS_LIMIT	NOPROTO	{ int cap_fcntls_limit(int fd, \
+				    uint32_t fcntlrights); }
+537	AUE_CAP_FCNTLS_GET	NOPROTO	{ int cap_fcntls_get(int fd, \
+				    uint32_t *fcntlrightsp); }
diff --git a/sys/compat/linux/linux_file.c b/sys/compat/linux/linux_file.c
index 0318a5c..346d178 100644
--- a/sys/compat/linux/linux_file.c
+++ b/sys/compat/linux/linux_file.c
@@ -154,6 +154,7 @@ linux_common_open(struct thread *td, int dirfd, char *path, int l_flags, int mod
 			SESS_LEADER(p) && !(p->p_flag & P_CONTROLT)) {
 			    PROC_UNLOCK(p);
 			    sx_unlock(&proctree_lock);
+			    /* XXXPJD: Verify if TIOCSCTTY is allowed. */
 			    if (fp->f_type == DTYPE_VNODE)
 				    (void) fo_ioctl(fp, TIOCSCTTY, (caddr_t) 0,
 					     td->td_ucred, td);
@@ -1038,11 +1039,11 @@ linux_pread(td, uap)
 	error = sys_pread(td, &bsd);
 
 	if (error == 0) {
-   	   	/* This seems to violate POSIX but linux does it */
-		if ((error = fgetvp(td, uap->fd, CAP_READ, &vp)) != 0)
-   		   	return (error);
+		/* This seems to violate POSIX but linux does it */
+		if ((error = fgetvp(td, uap->fd, CAP_PREAD, &vp)) != 0)
+			return (error);
 		if (vp->v_type == VDIR) {
-   		   	vrele(vp);
+			vrele(vp);
 			return (EISDIR);
 		}
 		vrele(vp);
diff --git a/sys/compat/svr4/svr4_fcntl.c b/sys/compat/svr4/svr4_fcntl.c
index b9d3ace..86fab78 100644
--- a/sys/compat/svr4/svr4_fcntl.c
+++ b/sys/compat/svr4/svr4_fcntl.c
@@ -265,14 +265,14 @@ fd_revoke(td, fd)
 	/*
 	 * If we ever want to support Capsicum on SVR4 processes (unlikely)
 	 * or FreeBSD grows a native frevoke() (more likely), we will need a
-	 * CAP_REVOKE here.
+	 * CAP_FREVOKE here.
 	 *
-	 * In the meantime, use CAP_MASK_VALID: if a SVR4 process wants to
+	 * 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).
 	 */
-	if ((error = fgetvp(td, fd, CAP_MASK_VALID, &vp)) != 0)
+	if ((error = fgetvp(td, fd, CAP_ALL, &vp)) != 0)
 		return (error);
 
 	if (vp->v_type != VCHR && vp->v_type != VBLK) {
diff --git a/sys/compat/svr4/svr4_filio.c b/sys/compat/svr4/svr4_filio.c
index 967169b..0fbba07 100644
--- a/sys/compat/svr4/svr4_filio.c
+++ b/sys/compat/svr4/svr4_filio.c
@@ -197,22 +197,24 @@ svr4_fil_ioctl(fp, td, retval, fd, cmd, data)
 	u_long cmd;
 	caddr_t data;
 {
-	int error;
-	int num;
 	struct filedesc *fdp = td->td_proc->p_fd;
+	struct filedescent *fde;
+	int error, num;
 
 	*retval = 0;
 
 	switch (cmd) {
 	case SVR4_FIOCLEX:
 		FILEDESC_XLOCK(fdp);
-		fdp->fd_ofileflags[fd] |= UF_EXCLOSE;
+		fde = &fdp->fd_ofiles[fd];
+		fde->fde_flags |= UF_EXCLOSE;
 		FILEDESC_XUNLOCK(fdp);
 		return 0;
 
 	case SVR4_FIONCLEX:
 		FILEDESC_XLOCK(fdp);
-		fdp->fd_ofileflags[fd] &= ~UF_EXCLOSE;
+		fde = &fdp->fd_ofiles[fd];
+		fde->fde_flags &= ~UF_EXCLOSE;
 		FILEDESC_XUNLOCK(fdp);
 		return 0;
 
diff --git a/sys/compat/svr4/svr4_misc.c b/sys/compat/svr4/svr4_misc.c
index d6bc4eb..0cfaeae 100644
--- a/sys/compat/svr4/svr4_misc.c
+++ b/sys/compat/svr4/svr4_misc.c
@@ -247,10 +247,8 @@ svr4_sys_getdents64(td, uap)
 
 	DPRINTF(("svr4_sys_getdents64(%d, *, %d)\n",
 		uap->fd, uap->nbytes));
-	if ((error = getvnode(td->td_proc->p_fd, uap->fd,
-	    CAP_READ | CAP_SEEK, &fp)) != 0) {
+	if ((error = getvnode(td->td_proc->p_fd, uap->fd, CAP_READ, &fp)) != 0)
 		return (error);
-	}
 
 	if ((fp->f_flag & FREAD) == 0) {
 		fdrop(fp, td);
@@ -426,8 +424,7 @@ svr4_sys_getdents(td, uap)
 	if (uap->nbytes < 0)
 		return (EINVAL);
 
-	if ((error = getvnode(td->td_proc->p_fd, uap->fd,
-	    CAP_READ | CAP_SEEK, &fp)) != 0)
+	if ((error = getvnode(td->td_proc->p_fd, uap->fd, CAP_READ, &fp)) != 0)
 		return (error);
 
 	if ((fp->f_flag & FREAD) == 0) {
diff --git a/sys/compat/svr4/svr4_stream.c b/sys/compat/svr4/svr4_stream.c
index 27014e3..1c7e83e 100644
--- a/sys/compat/svr4/svr4_stream.c
+++ b/sys/compat/svr4/svr4_stream.c
@@ -1449,7 +1449,7 @@ svr4_sys_putmsg(td, uap)
 	struct file     *fp;
 	int error;
 
-	if ((error = fget(td, uap->fd, CAP_WRITE, &fp)) != 0) {
+	if ((error = fget(td, uap->fd, CAP_SEND, &fp)) != 0) {
 #ifdef DEBUG_SVR4
 	        uprintf("putmsg: bad fp\n");
 #endif
@@ -1621,7 +1621,7 @@ svr4_sys_getmsg(td, uap)
 	struct file     *fp;
 	int error;
 
-	if ((error = fget(td, uap->fd, CAP_READ, &fp)) != 0) {
+	if ((error = fget(td, uap->fd, CAP_RECV, &fp)) != 0) {
 #ifdef DEBUG_SVR4
 	        uprintf("getmsg: bad fp\n");
 #endif