flag parameters: paccept

This patch is by far the most complex in the series.  It adds a new syscall
paccept.  This syscall differs from accept in that it adds (at the userlevel)
two additional parameters:

- a signal mask
- a flags value

The flags parameter can be used to set flag like SOCK_CLOEXEC.  This is
imlpemented here as well.  Some people argued that this is a property which
should be inherited from the file desriptor for the server but this is against
POSIX.  Additionally, we really want the signal mask parameter as well
(similar to pselect, ppoll, etc).  So an interface change in inevitable.

The flag value is the same as for socket and socketpair.  I think diverging
here will only create confusion.  Similar to the filesystem interfaces where
the use of the O_* constants differs, it is acceptable here.

The signal mask is handled as for pselect etc.  The mask is temporarily
installed for the thread and removed before the call returns.  I modeled the
code after pselect.  If there is a problem it's likely also in pselect.

For architectures which use socketcall I maintained this interface instead of
adding a system call.  The symmetry shouldn't be broken.

The following test must be adjusted for architectures other than x86 and
x86-64 and in case the syscall numbers changed.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#include <errno.h>
#include <fcntl.h>
#include <pthread.h>
#include <signal.h>
#include <stdio.h>
#include <unistd.h>
#include <netinet/in.h>
#include <sys/socket.h>
#include <sys/syscall.h>

#ifndef __NR_paccept
# ifdef __x86_64__
#  define __NR_paccept 288
# elif defined __i386__
#  define SYS_PACCEPT 18
#  define USE_SOCKETCALL 1
# else
#  error "need __NR_paccept"
# endif
#endif

#ifdef USE_SOCKETCALL
# define paccept(fd, addr, addrlen, mask, flags) \
  ({ long args[6] = { \
       (long) fd, (long) addr, (long) addrlen, (long) mask, 8, (long) flags }; \
     syscall (__NR_socketcall, SYS_PACCEPT, args); })
#else
# define paccept(fd, addr, addrlen, mask, flags) \
  syscall (__NR_paccept, fd, addr, addrlen, mask, 8, flags)
#endif

#define PORT 57392

#define SOCK_CLOEXEC O_CLOEXEC

static pthread_barrier_t b;

static void *
tf (void *arg)
{
  pthread_barrier_wait (&b);
  int s = socket (AF_INET, SOCK_STREAM, 0);
  struct sockaddr_in sin;
  sin.sin_family = AF_INET;
  sin.sin_addr.s_addr = htonl (INADDR_LOOPBACK);
  sin.sin_port = htons (PORT);
  connect (s, (const struct sockaddr *) &sin, sizeof (sin));
  close (s);

  pthread_barrier_wait (&b);
  s = socket (AF_INET, SOCK_STREAM, 0);
  sin.sin_port = htons (PORT);
  connect (s, (const struct sockaddr *) &sin, sizeof (sin));
  close (s);
  pthread_barrier_wait (&b);

  pthread_barrier_wait (&b);
  sleep (2);
  pthread_kill ((pthread_t) arg, SIGUSR1);

  return NULL;
}

static void
handler (int s)
{
}

int
main (void)
{
  pthread_barrier_init (&b, NULL, 2);

  struct sockaddr_in sin;
  pthread_t th;
  if (pthread_create (&th, NULL, tf, (void *) pthread_self ()) != 0)
    {
      puts ("pthread_create failed");
      return 1;
    }

  int s = socket (AF_INET, SOCK_STREAM, 0);
  int reuse = 1;
  setsockopt (s, SOL_SOCKET, SO_REUSEADDR, &reuse, sizeof (reuse));
  sin.sin_family = AF_INET;
  sin.sin_addr.s_addr = htonl (INADDR_LOOPBACK);
  sin.sin_port = htons (PORT);
  bind (s, (struct sockaddr *) &sin, sizeof (sin));
  listen (s, SOMAXCONN);

  pthread_barrier_wait (&b);

  int s2 = paccept (s, NULL, 0, NULL, 0);
  if (s2 < 0)
    {
      puts ("paccept(0) failed");
      return 1;
    }

  int coe = fcntl (s2, F_GETFD);
  if (coe & FD_CLOEXEC)
    {
      puts ("paccept(0) set close-on-exec-flag");
      return 1;
    }
  close (s2);

  pthread_barrier_wait (&b);

  s2 = paccept (s, NULL, 0, NULL, SOCK_CLOEXEC);
  if (s2 < 0)
    {
      puts ("paccept(SOCK_CLOEXEC) failed");
      return 1;
    }

  coe = fcntl (s2, F_GETFD);
  if ((coe & FD_CLOEXEC) == 0)
    {
      puts ("paccept(SOCK_CLOEXEC) does not set close-on-exec flag");
      return 1;
    }
  close (s2);

  pthread_barrier_wait (&b);

  struct sigaction sa;
  sa.sa_handler = handler;
  sa.sa_flags = 0;
  sigemptyset (&sa.sa_mask);
  sigaction (SIGUSR1, &sa, NULL);

  sigset_t ss;
  pthread_sigmask (SIG_SETMASK, NULL, &ss);
  sigaddset (&ss, SIGUSR1);
  pthread_sigmask (SIG_SETMASK, &ss, NULL);

  sigdelset (&ss, SIGUSR1);
  alarm (4);
  pthread_barrier_wait (&b);

  errno = 0 ;
  s2 = paccept (s, NULL, 0, &ss, 0);
  if (s2 != -1 || errno != EINTR)
    {
      puts ("paccept did not fail with EINTR");
      return 1;
    }

  close (s);

  puts ("OK");

  return 0;
}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[akpm@linux-foundation.org: make it compile]
[akpm@linux-foundation.org: add sys_ni stub]
Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Cc: <linux-arch@vger.kernel.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Roland McGrath <roland@redhat.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

8 files changed, 139 insertions, 12 deletions

diff --git a/include/asm-alpha/socket.h b/include/asm-alpha/socket.h
index 08c979319929..a1057c2d95e7 100644
--- a/include/asm-alpha/socket.h
+++ b/include/asm-alpha/socket.h
@@ -62,4 +62,9 @@
 
 #define SO_MARK                 36
 
+/* O_NONBLOCK clashes with the bits used for socket types.  Therefore we
+ * have to define SOCK_NONBLOCK to a different value here.
+ */
+#define SOCK_NONBLOCK   0x40000000
+
 #endif /* _ASM_SOCKET_H */
diff --git a/include/asm-parisc/socket.h b/include/asm-parisc/socket.h
index 69a7a0d30b02..fba402c95ac2 100644
--- a/include/asm-parisc/socket.h
+++ b/include/asm-parisc/socket.h
@@ -54,4 +54,9 @@
 
 #define SO_MARK                 0x401f
 
+/* O_NONBLOCK clashes with the bits used for socket types.  Therefore we
+ * have to define SOCK_NONBLOCK to a different value here.
+ */
+#define SOCK_NONBLOCK   0x40000000
+
 #endif /* _ASM_SOCKET_H */
diff --git a/include/asm-x86/unistd_64.h b/include/asm-x86/unistd_64.h
index 9c1a4a3470d9..e323994a370f 100644
--- a/include/asm-x86/unistd_64.h
+++ b/include/asm-x86/unistd_64.h
@@ -639,6 +639,8 @@ __SYSCALL(__NR_fallocate, sys_fallocate)
 __SYSCALL(__NR_timerfd_settime, sys_timerfd_settime)
 #define __NR_timerfd_gettime                    287
 __SYSCALL(__NR_timerfd_gettime, sys_timerfd_gettime)
+#define __NR_paccept                            288
+__SYSCALL(__NR_paccept, sys_paccept)
 
 
 #ifndef __NO_STUBS
diff --git a/include/linux/net.h b/include/linux/net.h
index 8b5383c45b45..3a9b06d4d0fe 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -47,6 +47,7 @@ struct net;
 #define SYS_GETSOCKOPT  15              /* sys_getsockopt(2)            */
 #define SYS_SENDMSG     16              /* sys_sendmsg(2)               */
 #define SYS_RECVMSG     17              /* sys_recvmsg(2)               */
+#define SYS_PACCEPT     18              /* sys_paccept(2)               */
 
 typedef enum {
         SS_FREE = 0,                    /* not allocated                */
@@ -219,6 +220,8 @@ extern int 	     sock_map_fd(struct socket *sock, int flags);
 extern struct socket *sockfd_lookup(int fd, int *err);
 #define              sockfd_put(sock) fput(sock->file)
 extern int           net_ratelimit(void);
+extern long          do_accept(int fd, struct sockaddr __user *upeer_sockaddr,
+                               int __user *upeer_addrlen, int flags);
 
 #define net_random()            random32()
 #define net_srandom(seed)       srandom32((__force u32)seed)
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 4394dadff813..2a2a40af6b2c 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -409,6 +409,8 @@ asmlinkage long sys_getsockopt(int fd, int level, int optname,
 asmlinkage long sys_bind(int, struct sockaddr __user *, int);
 asmlinkage long sys_connect(int, struct sockaddr __user *, int);
 asmlinkage long sys_accept(int, struct sockaddr __user *, int __user *);
+asmlinkage long sys_paccept(int, struct sockaddr __user *, int __user *,
+                            const sigset_t *, size_t, int);
 asmlinkage long sys_getsockname(int, struct sockaddr __user *, int __user *);
 asmlinkage long sys_getpeername(int, struct sockaddr __user *, int __user *);
 asmlinkage long sys_send(int, void __user *, size_t, unsigned);
diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
index 0fea0ee12da9..2f0b8a2e600f 100644
--- a/kernel/sys_ni.c
+++ b/kernel/sys_ni.c
@@ -31,6 +31,7 @@ cond_syscall(sys_socketpair);
 cond_syscall(sys_bind);
 cond_syscall(sys_listen);
 cond_syscall(sys_accept);
+cond_syscall(sys_paccept);
 cond_syscall(sys_connect);
 cond_syscall(sys_getsockname);
 cond_syscall(sys_getpeername);
diff --git a/net/compat.c b/net/compat.c
index 6e1b03b51933..67fb6a3834a3 100644
--- a/net/compat.c
+++ b/net/compat.c
@@ -722,9 +722,10 @@ EXPORT_SYMBOL(compat_mc_getsockopt);
 
 /* Argument list sizes for compat_sys_socketcall */
 #define AL(x) ((x) * sizeof(u32))
-static unsigned char nas[18]={AL(0),AL(3),AL(3),AL(3),AL(2),AL(3),
+static unsigned char nas[19]={AL(0),AL(3),AL(3),AL(3),AL(2),AL(3),
                                 AL(3),AL(3),AL(4),AL(4),AL(4),AL(6),
-                                AL(6),AL(2),AL(5),AL(5),AL(3),AL(3)};
+                                AL(6),AL(2),AL(5),AL(5),AL(3),AL(3),
+                                AL(6)};
 #undef AL
 
 asmlinkage long compat_sys_sendmsg(int fd, struct compat_msghdr __user *msg, unsigned flags)
@@ -737,13 +738,52 @@ asmlinkage long compat_sys_recvmsg(int fd, struct compat_msghdr __user *msg, uns
         return sys_recvmsg(fd, (struct msghdr __user *)msg, flags | MSG_CMSG_COMPAT);
 }
 
+asmlinkage long compat_sys_paccept(int fd, struct sockaddr __user *upeer_sockaddr,
+                                   int __user *upeer_addrlen,
+                                   const compat_sigset_t __user *sigmask,
+                                   compat_size_t sigsetsize, int flags)
+{
+        compat_sigset_t ss32;
+        sigset_t ksigmask, sigsaved;
+        int ret;
+
+        if (sigmask) {
+                if (sigsetsize != sizeof(compat_sigset_t))
+                        return -EINVAL;
+                if (copy_from_user(&ss32, sigmask, sizeof(ss32)))
+                        return -EFAULT;
+                sigset_from_compat(&ksigmask, &ss32);
+
+                sigdelsetmask(&ksigmask, sigmask(SIGKILL)|sigmask(SIGSTOP));
+                sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved);
+        }
+
+        ret = do_accept(fd, upeer_sockaddr, upeer_addrlen, flags);
+
+        if (ret == -ERESTARTNOHAND) {
+                /*
+                 * Don't restore the signal mask yet. Let do_signal() deliver
+                 * the signal on the way back to userspace, before the signal
+                 * mask is restored.
+                 */
+                if (sigmask) {
+                        memcpy(&current->saved_sigmask, &sigsaved,
+                               sizeof(sigsaved));
+                        set_restore_sigmask();
+                }
+        } else if (sigmask)
+                sigprocmask(SIG_SETMASK, &sigsaved, NULL);
+
+        return ret;
+}
+
 asmlinkage long compat_sys_socketcall(int call, u32 __user *args)
 {
         int ret;
         u32 a[6];
         u32 a0, a1;
 
-        if (call < SYS_SOCKET || call > SYS_RECVMSG)
+        if (call < SYS_SOCKET || call > SYS_PACCEPT)
                 return -EINVAL;
         if (copy_from_user(a, args, nas[call]))
                 return -EFAULT;
@@ -764,7 +804,7 @@ asmlinkage long compat_sys_socketcall(int call, u32 __user *args)
                 ret = sys_listen(a0, a1);
                 break;
         case SYS_ACCEPT:
-                ret = sys_accept(a0, compat_ptr(a1), compat_ptr(a[2]));
+                ret = do_accept(a0, compat_ptr(a1), compat_ptr(a[2]), 0);
                 break;
         case SYS_GETSOCKNAME:
                 ret = sys_getsockname(a0, compat_ptr(a1), compat_ptr(a[2]));
@@ -804,6 +844,10 @@ asmlinkage long compat_sys_socketcall(int call, u32 __user *args)
         case SYS_RECVMSG:
                 ret = compat_sys_recvmsg(a0, compat_ptr(a1), a[2]);
                 break;
+        case SYS_PACCEPT:
+                ret = compat_sys_paccept(a0, compat_ptr(a1), compat_ptr(a[2]),
+                                         compat_ptr(a[3]), a[4], a[5]);
+                break;
         default:
                 ret = -EINVAL;
                 break;
diff --git a/net/socket.c b/net/socket.c
index 64601f900352..a0ce8ad72252 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -63,6 +63,7 @@
 #include <linux/file.h>
 #include <linux/net.h>
 #include <linux/interrupt.h>
+#include <linux/thread_info.h>
 #include <linux/rcupdate.h>
 #include <linux/netdevice.h>
 #include <linux/proc_fs.h>
@@ -1225,6 +1226,9 @@ asmlinkage long sys_socket(int family, int type, int protocol)
                 return -EINVAL;
         type &= SOCK_TYPE_MASK;
 
+        if (SOCK_NONBLOCK != O_NONBLOCK && (flags & SOCK_NONBLOCK))
+                flags = (flags & ~SOCK_NONBLOCK) | O_NONBLOCK;
+
         retval = sock_create(family, type, protocol, &sock);
         if (retval < 0)
                 goto out;
@@ -1259,6 +1263,9 @@ asmlinkage long sys_socketpair(int family, int type, int protocol,
                 return -EINVAL;
         type &= SOCK_TYPE_MASK;
 
+        if (SOCK_NONBLOCK != O_NONBLOCK && (flags & SOCK_NONBLOCK))
+                flags = (flags & ~SOCK_NONBLOCK) | O_NONBLOCK;
+
         /*
          * Obtain the first socket and check if the underlying protocol
          * supports the socketpair call.
@@ -1413,14 +1420,20 @@ asmlinkage long sys_listen(int fd, int backlog)
  *      clean when we restucture accept also.
  */
 
-asmlinkage long sys_accept(int fd, struct sockaddr __user *upeer_sockaddr,
-                           int __user *upeer_addrlen)
+long do_accept(int fd, struct sockaddr __user *upeer_sockaddr,
+               int __user *upeer_addrlen, int flags)
 {
         struct socket *sock, *newsock;
         struct file *newfile;
         int err, len, newfd, fput_needed;
         struct sockaddr_storage address;
 
+        if (flags & ~SOCK_CLOEXEC)
+                return -EINVAL;
+
+        if (SOCK_NONBLOCK != O_NONBLOCK && (flags & SOCK_NONBLOCK))
+                flags = (flags & ~SOCK_NONBLOCK) | O_NONBLOCK;
+
         sock = sockfd_lookup_light(fd, &err, &fput_needed);
         if (!sock)
                 goto out;
@@ -1438,7 +1451,7 @@ asmlinkage long sys_accept(int fd, struct sockaddr __user *upeer_sockaddr,
          */
         __module_get(newsock->ops->owner);
 
-        newfd = sock_alloc_fd(&newfile, 0);
+        newfd = sock_alloc_fd(&newfile, flags & O_CLOEXEC);
         if (unlikely(newfd < 0)) {
                 err = newfd;
                 sock_release(newsock);
@@ -1491,6 +1504,50 @@ out_fd:
         goto out_put;
 }
 
+asmlinkage long sys_paccept(int fd, struct sockaddr __user *upeer_sockaddr,
+                            int __user *upeer_addrlen,
+                            const sigset_t __user *sigmask,
+                            size_t sigsetsize, int flags)
+{
+        sigset_t ksigmask, sigsaved;
+        int ret;
+
+        if (sigmask) {
+                /* XXX: Don't preclude handling different sized sigset_t's.  */
+                if (sigsetsize != sizeof(sigset_t))
+                        return -EINVAL;
+                if (copy_from_user(&ksigmask, sigmask, sizeof(ksigmask)))
+                        return -EFAULT;
+
+                sigdelsetmask(&ksigmask, sigmask(SIGKILL)|sigmask(SIGSTOP));
+                sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved);
+        }
+
+        ret = do_accept(fd, upeer_sockaddr, upeer_addrlen, flags);
+
+        if (ret < 0 && signal_pending(current)) {
+                /*
+                 * Don't restore the signal mask yet. Let do_signal() deliver
+                 * the signal on the way back to userspace, before the signal
+                 * mask is restored.
+                 */
+                if (sigmask) {
+                        memcpy(&current->saved_sigmask, &sigsaved,
+                               sizeof(sigsaved));
+                        set_restore_sigmask();
+                }
+        } else if (sigmask)
+                sigprocmask(SIG_SETMASK, &sigsaved, NULL);
+
+        return ret;
+}
+
+asmlinkage long sys_accept(int fd, struct sockaddr __user *upeer_sockaddr,
+                           int __user *upeer_addrlen)
+{
+        return do_accept(fd, upeer_sockaddr, upeer_addrlen, 0);
+}
+
 /*
  *      Attempt to connect to a socket with the server address.  The address
  *      is in user space so we verify it is OK and move it to kernel space.
@@ -2011,10 +2068,11 @@ out:
 
 /* Argument list sizes for sys_socketcall */
 #define AL(x) ((x) * sizeof(unsigned long))
-static const unsigned char nargs[18]={
+static const unsigned char nargs[19]={
         AL(0),AL(3),AL(3),AL(3),AL(2),AL(3),
         AL(3),AL(3),AL(4),AL(4),AL(4),AL(6),
-        AL(6),AL(2),AL(5),AL(5),AL(3),AL(3)
+        AL(6),AL(2),AL(5),AL(5),AL(3),AL(3),
+        AL(6)
 };
 
 #undef AL
@@ -2033,7 +2091,7 @@ asmlinkage long sys_socketcall(int call, unsigned long __user *args)
         unsigned long a0, a1;
         int err;
 
-        if (call < 1 || call > SYS_RECVMSG)
+        if (call < 1 || call > SYS_PACCEPT)
                 return -EINVAL;
 
         /* copy_from_user should be SMP safe. */
@@ -2062,8 +2120,8 @@ asmlinkage long sys_socketcall(int call, unsigned long __user *args)
                 break;
         case SYS_ACCEPT:
                 err =
-                    sys_accept(a0, (struct sockaddr __user *)a1,
-                               (int __user *)a[2]);
+                    do_accept(a0, (struct sockaddr __user *)a1,
+                              (int __user *)a[2], 0);
                 break;
         case SYS_GETSOCKNAME:
                 err =
@@ -2110,6 +2168,13 @@ asmlinkage long sys_socketcall(int call, unsigned long __user *args)
         case SYS_RECVMSG:
                 err = sys_recvmsg(a0, (struct msghdr __user *)a1, a[2]);
                 break;
+        case SYS_PACCEPT:
+                err =
+                    sys_paccept(a0, (struct sockaddr __user *)a1,
+                                (int __user *)a[2],
+                                (const sigset_t __user *) a[3],
+                                a[4], a[5]);
+                break;
         default:
                 err = -EINVAL;
                 break;