Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6:
  [PATCH] fix sysctl_nr_open bugs
  [PATCH] sanitize anon_inode_getfd()
  [PATCH] split linux/file.h
  [PATCH] make osf_select() use core_sys_select()
  [PATCH] remove horrors with irix tty ioctls handling
  [PATCH] fix file and descriptor handling in perfmon

16 files changed, 55 insertions, 60 deletions

diff --git a/fs/anon_inodes.c b/fs/anon_inodes.c
index f42be069e085..977ef208c051 100644
--- a/fs/anon_inodes.c
+++ b/fs/anon_inodes.c
@@ -57,9 +57,6 @@ static struct dentry_operations anon_inodefs_dentry_operations = {
  *                    anonymous inode, and a dentry that describe the "class"
  *                    of the file
  *
- * @pfd:     [out]   pointer to the file descriptor
- * @dpinode: [out]   pointer to the inode
- * @pfile:   [out]   pointer to the file struct
  * @name:    [in]    name of the "class" of the new file
  * @fops     [in]    file operations for the new file
  * @priv     [in]    private data for the new file (will be file's private_data)
@@ -68,10 +65,9 @@ static struct dentry_operations anon_inodefs_dentry_operations = {
  * that do not need to have a full-fledged inode in order to operate correctly.
  * All the files created with anon_inode_getfd() will share a single inode,
  * hence saving memory and avoiding code duplication for the file/inode/dentry
- * setup.
+ * setup.  Returns new descriptor or -error.
  */
-int anon_inode_getfd(int *pfd, struct inode **pinode, struct file **pfile,
-                     const char *name, const struct file_operations *fops,
+int anon_inode_getfd(const char *name, const struct file_operations *fops,
                      void *priv)
 {
         struct qstr this;
@@ -125,10 +121,7 @@ int anon_inode_getfd(int *pfd, struct inode **pinode, struct file **pfile,
 
         fd_install(fd, file);
 
-        *pfd = fd;
-        *pinode = anon_inode_inode;
-        *pfile = file;
-        return 0;
+        return fd;
 
 err_dput:
         dput(dentry);
diff --git a/fs/compat.c b/fs/compat.c
index 139dc93c092d..332a869d2c53 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -24,6 +24,7 @@
 #include <linux/fcntl.h>
 #include <linux/namei.h>
 #include <linux/file.h>
+#include <linux/fdtable.h>
 #include <linux/vfs.h>
 #include <linux/ioctl.h>
 #include <linux/init.h>
diff --git a/fs/dnotify.c b/fs/dnotify.c
index eaecc4cfe540..676073b8dda5 100644
--- a/fs/dnotify.c
+++ b/fs/dnotify.c
@@ -20,7 +20,7 @@
 #include <linux/init.h>
 #include <linux/spinlock.h>
 #include <linux/slab.h>
-#include <linux/file.h>
+#include <linux/fdtable.h>
 
 int dir_notify_enable __read_mostly = 1;
 
diff --git a/fs/eventfd.c b/fs/eventfd.c
index a9f130cd50ac..343942deeec1 100644
--- a/fs/eventfd.c
+++ b/fs/eventfd.c
@@ -200,10 +200,8 @@ struct file *eventfd_fget(int fd)
 
 asmlinkage long sys_eventfd(unsigned int count)
 {
-        int error, fd;
+        int fd;
         struct eventfd_ctx *ctx;
-        struct file *file;
-        struct inode *inode;
 
         ctx = kmalloc(sizeof(*ctx), GFP_KERNEL);
         if (!ctx)
@@ -216,12 +214,9 @@ asmlinkage long sys_eventfd(unsigned int count)
          * When we call this, the initialization must be complete, since
          * anon_inode_getfd() will install the fd.
          */
-        error = anon_inode_getfd(&fd, &inode, &file, "[eventfd]",
-                                 &eventfd_fops, ctx);
-        if (!error)
-                return fd;
-
-        kfree(ctx);
-        return error;
+        fd = anon_inode_getfd("[eventfd]", &eventfd_fops, ctx);
+        if (fd < 0)
+                kfree(ctx);
+        return fd;
 }
 
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 221086fef174..990c01d2d66b 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -1050,8 +1050,6 @@ asmlinkage long sys_epoll_create(int size)
 {
         int error, fd = -1;
         struct eventpoll *ep;
-        struct inode *inode;
-        struct file *file;
 
         DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_create(%d)\n",
                      current, size));
@@ -1061,29 +1059,24 @@ asmlinkage long sys_epoll_create(int size)
          * structure ( "struct eventpoll" ).
          */
         error = -EINVAL;
-        if (size <= 0 || (error = ep_alloc(&ep)) != 0)
+        if (size <= 0 || (error = ep_alloc(&ep)) < 0) {
+                fd = error;
                 goto error_return;
+        }
 
         /*
          * Creates all the items needed to setup an eventpoll file. That is,
-         * a file structure, and inode and a free file descriptor.
+         * a file structure and a free file descriptor.
          */
-        error = anon_inode_getfd(&fd, &inode, &file, "[eventpoll]",
-                                 &eventpoll_fops, ep);
-        if (error)
-                goto error_free;
+        fd = anon_inode_getfd("[eventpoll]", &eventpoll_fops, ep);
+        if (fd < 0)
+                ep_free(ep);
 
+error_return:
         DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_create(%d) = %d\n",
                      current, size, fd));
 
         return fd;
-
-error_free:
-        ep_free(ep);
-error_return:
-        DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_create(%d) = %d\n",
-                     current, size, error));
-        return error;
 }
 
 /*
diff --git a/fs/exec.c b/fs/exec.c
index 9f9f931ef949..aeaa9791d8be 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -24,6 +24,7 @@
 
 #include <linux/slab.h>
 #include <linux/file.h>
+#include <linux/fdtable.h>
 #include <linux/mman.h>
 #include <linux/a.out.h>
 #include <linux/stat.h>
diff --git a/fs/fcntl.c b/fs/fcntl.c
index 3f3ac630ccde..bfd776509a72 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -9,6 +9,7 @@
 #include <linux/mm.h>
 #include <linux/fs.h>
 #include <linux/file.h>
+#include <linux/fdtable.h>
 #include <linux/capability.h>
 #include <linux/dnotify.h>
 #include <linux/smp_lock.h>
diff --git a/fs/file.c b/fs/file.c
index 5110acb1c9ef..4c6f0ea12c41 100644
--- a/fs/file.c
+++ b/fs/file.c
@@ -12,6 +12,7 @@
 #include <linux/slab.h>
 #include <linux/vmalloc.h>
 #include <linux/file.h>
+#include <linux/fdtable.h>
 #include <linux/bitops.h>
 #include <linux/interrupt.h>
 #include <linux/spinlock.h>
@@ -149,8 +150,16 @@ static struct fdtable * alloc_fdtable(unsigned int nr)
         nr /= (1024 / sizeof(struct file *));
         nr = roundup_pow_of_two(nr + 1);
         nr *= (1024 / sizeof(struct file *));
-        if (nr > sysctl_nr_open)
-                nr = sysctl_nr_open;
+        /*
+         * Note that this can drive nr *below* what we had passed if sysctl_nr_open
+         * had been set lower between the check in expand_files() and here.  Deal
+         * with that in caller, it's cheaper that way.
+         *
+         * We make sure that nr remains a multiple of BITS_PER_LONG - otherwise
+         * bitmaps handling below becomes unpleasant, to put it mildly...
+         */
+        if (unlikely(nr > sysctl_nr_open))
+                nr = ((sysctl_nr_open - 1) | (BITS_PER_LONG - 1)) + 1;
 
         fdt = kmalloc(sizeof(struct fdtable), GFP_KERNEL);
         if (!fdt)
@@ -199,6 +208,16 @@ static int expand_fdtable(struct files_struct *files, int nr)
         if (!new_fdt)
                 return -ENOMEM;
         /*
+         * extremely unlikely race - sysctl_nr_open decreased between the check in
+         * caller and alloc_fdtable().  Cheaper to catch it here...
+         */
+        if (unlikely(new_fdt->max_fds <= nr)) {
+                free_fdarr(new_fdt);
+                free_fdset(new_fdt);
+                kfree(new_fdt);
+                return -EMFILE;
+        }
+        /*
          * Check again since another task may have expanded the fd table while
          * we dropped the lock
          */
diff --git a/fs/file_table.c b/fs/file_table.c
index 7a0a9b872251..83084225b4c3 100644
--- a/fs/file_table.c
+++ b/fs/file_table.c
@@ -8,6 +8,7 @@
 #include <linux/string.h>
 #include <linux/slab.h>
 #include <linux/file.h>
+#include <linux/fdtable.h>
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/fs.h>
diff --git a/fs/locks.c b/fs/locks.c
index 44d9a6a7ec50..663c069b59b3 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -116,6 +116,7 @@
 
 #include <linux/capability.h>
 #include <linux/file.h>
+#include <linux/fdtable.h>
 #include <linux/fs.h>
 #include <linux/init.h>
 #include <linux/module.h>
diff --git a/fs/open.c b/fs/open.c
index 7af1f05d5978..a1450086e92f 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -7,6 +7,7 @@
 #include <linux/string.h>
 #include <linux/mm.h>
 #include <linux/file.h>
+#include <linux/fdtable.h>
 #include <linux/quotaops.h>
 #include <linux/fsnotify.h>
 #include <linux/module.h>
diff --git a/fs/proc/array.c b/fs/proc/array.c
index c135cbdd9127..dca997a93bff 100644
--- a/fs/proc/array.c
+++ b/fs/proc/array.c
@@ -73,6 +73,7 @@
 #include <linux/signal.h>
 #include <linux/highmem.h>
 #include <linux/file.h>
+#include <linux/fdtable.h>
 #include <linux/times.h>
 #include <linux/cpuset.h>
 #include <linux/rcupdate.h>
diff --git a/fs/proc/base.c b/fs/proc/base.c
index fcf02f2deeba..808cbdc193d3 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -56,6 +56,7 @@
 #include <linux/init.h>
 #include <linux/capability.h>
 #include <linux/file.h>
+#include <linux/fdtable.h>
 #include <linux/string.h>
 #include <linux/seq_file.h>
 #include <linux/namei.h>
diff --git a/fs/select.c b/fs/select.c
index 2c292146e246..8dda969614a9 100644
--- a/fs/select.c
+++ b/fs/select.c
@@ -21,6 +21,7 @@
 #include <linux/poll.h>
 #include <linux/personality.h> /* for STICKY_TIMEOUTS */
 #include <linux/file.h>
+#include <linux/fdtable.h>
 #include <linux/fs.h>
 #include <linux/rcupdate.h>
 
@@ -298,7 +299,7 @@ int do_select(int n, fd_set_bits *fds, s64 *timeout)
 #define MAX_SELECT_SECONDS \
         ((unsigned long) (MAX_SCHEDULE_TIMEOUT / HZ)-1)
 
-static int core_sys_select(int n, fd_set __user *inp, fd_set __user *outp,
+int core_sys_select(int n, fd_set __user *inp, fd_set __user *outp,
                            fd_set __user *exp, s64 *timeout)
 {
         fd_set_bits fds;
diff --git a/fs/signalfd.c b/fs/signalfd.c
index 8ead0db35933..619725644c75 100644
--- a/fs/signalfd.c
+++ b/fs/signalfd.c
@@ -207,11 +207,8 @@ static const struct file_operations signalfd_fops = {
 
 asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, size_t sizemask)
 {
-        int error;
         sigset_t sigmask;
         struct signalfd_ctx *ctx;
-        struct file *file;
-        struct inode *inode;
 
         if (sizemask != sizeof(sigset_t) ||
             copy_from_user(&sigmask, user_mask, sizeof(sigmask)))
@@ -230,12 +227,11 @@ asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, size_t sizemas
                  * When we call this, the initialization must be complete, since
                  * anon_inode_getfd() will install the fd.
                  */
-                error = anon_inode_getfd(&ufd, &inode, &file, "[signalfd]",
-                                         &signalfd_fops, ctx);
-                if (error)
-                        goto err_fdalloc;
+                ufd = anon_inode_getfd("[signalfd]", &signalfd_fops, ctx);
+                if (ufd < 0)
+                        kfree(ctx);
         } else {
-                file = fget(ufd);
+                struct file *file = fget(ufd);
                 if (!file)
                         return -EBADF;
                 ctx = file->private_data;
@@ -252,9 +248,4 @@ asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, size_t sizemas
         }
 
         return ufd;
-
-err_fdalloc:
-        kfree(ctx);
-        return error;
 }
-
diff --git a/fs/timerfd.c b/fs/timerfd.c
index 5400524e9cb1..d87d354ec424 100644
--- a/fs/timerfd.c
+++ b/fs/timerfd.c
@@ -181,10 +181,8 @@ static struct file *timerfd_fget(int fd)
 
 asmlinkage long sys_timerfd_create(int clockid, int flags)
 {
-        int error, ufd;
+        int ufd;
         struct timerfd_ctx *ctx;
-        struct file *file;
-        struct inode *inode;
 
         if (flags)
                 return -EINVAL;
@@ -200,12 +198,9 @@ asmlinkage long sys_timerfd_create(int clockid, int flags)
         ctx->clockid = clockid;
         hrtimer_init(&ctx->tmr, clockid, HRTIMER_MODE_ABS);
 
-        error = anon_inode_getfd(&ufd, &inode, &file, "[timerfd]",
-                                 &timerfd_fops, ctx);
-        if (error) {
+        ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx);
+        if (ufd < 0)
                 kfree(ctx);
-                return error;
-        }
 
         return ufd;
 }