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] return to old errno choice in mkdir() et.al.
  [Patch] fs/binfmt_elf.c: fix wrong return values
  [PATCH] get rid of leak in compat_execve()
  [Patch] fs/binfmt_elf.c: fix a wrong free
  [PATCH] avoid multiplication overflows and signedness issues for max_fds
  [PATCH] dup_fd() part 4 - race fix
  [PATCH] dup_fd() - part 3
  [PATCH] dup_fd() part 2
  [PATCH] dup_fd() fixes, part 1
  [PATCH] take init_files to fs/file.c

arch/alpha/kernel/init_task.c

@@ -9,7 +9,6 @@
 
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);

arch/arm/kernel/init_task.c

@@ -13,7 +13,6 @@
 #include <asm/pgtable.h>
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);

arch/avr32/kernel/init_task.c

@@ -14,7 +14,6 @@
 #include <asm/pgtable.h>
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);

arch/blackfin/kernel/init_task.c

@@ -34,7 +34,6 @@
 #include <linux/fs.h>
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 

arch/cris/kernel/process.c

@@ -38,7 +38,6 @@
  */
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);

arch/frv/kernel/init_task.c

@@ -11,7 +11,6 @@
 
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);

arch/h8300/kernel/init_task.c

@@ -13,7 +13,6 @@
 #include <asm/pgtable.h>
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);

arch/ia64/kernel/init_task.c

@@ -18,7 +18,6 @@
 #include <asm/pgtable.h>
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);

arch/m32r/kernel/init_task.c

@@ -12,7 +12,6 @@
 #include <asm/pgtable.h>
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);

arch/m68k/kernel/process.c

@@ -41,7 +41,6 @@
  * setup.
  */
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);

arch/m68knommu/kernel/init_task.c

@@ -13,7 +13,6 @@
 #include <asm/pgtable.h>
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);

arch/mips/kernel/init_task.c

@@ -10,7 +10,6 @@
 #include <asm/pgtable.h>
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);

arch/mn10300/kernel/init_task.c

@@ -19,7 +19,6 @@
 #include <asm/pgtable.h>
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);

arch/parisc/kernel/init_task.c

@@ -35,7 +35,6 @@
 #include <asm/pgalloc.h>
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);

arch/powerpc/kernel/init_task.c

@@ -8,7 +8,6 @@
 #include <asm/uaccess.h>
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);

arch/s390/kernel/init_task.c

@@ -17,7 +17,6 @@
 #include <asm/pgtable.h>
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);

arch/sh/kernel/init_task.c

@@ -8,7 +8,6 @@
 #include <asm/pgtable.h>
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct pt_regs fake_swapper_regs;

arch/sparc/kernel/init_task.c

@@ -9,7 +9,6 @@
 #include <asm/uaccess.h>
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);

arch/sparc64/kernel/init_task.c

@@ -10,7 +10,6 @@
 #include <asm/processor.h>
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);

arch/um/kernel/init_task.c

@@ -12,7 +12,6 @@
 
 static struct fs_struct init_fs = INIT_FS;
 struct mm_struct init_mm = INIT_MM(init_mm);
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 EXPORT_SYMBOL(init_mm);

arch/v850/kernel/init_task.c

@@ -21,7 +21,6 @@
 #include <asm/pgtable.h>
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS (init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM (init_mm);

arch/x86/kernel/init_task.c

@@ -11,7 +11,6 @@
 #include <asm/desc.h>
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);

arch/xtensa/kernel/init_task.c

@@ -22,7 +22,6 @@
 #include <asm/uaccess.h>
 
 static struct fs_struct init_fs = INIT_FS;
-static struct files_struct init_files = INIT_FILES;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);

fs/binfmt_elf.c

@@ -256,7 +256,7 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
 			return -EFAULT;
 		len = strnlen_user((void __user *)p, MAX_ARG_STRLEN);
 		if (!len || len > MAX_ARG_STRLEN)
-			return 0;
+			return -EINVAL;
 		p += len;
 	}
 	if (__put_user(0, argv))
@@ -268,7 +268,7 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
 			return -EFAULT;
 		len = strnlen_user((void __user *)p, MAX_ARG_STRLEN);
 		if (!len || len > MAX_ARG_STRLEN)
-			return 0;
+			return -EINVAL;
 		p += len;
 	}
 	if (__put_user(0, envp))
@@ -1900,7 +1900,7 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file, un
 	/* alloc memory for large data structures: too large to be on stack */
 	elf = kmalloc(sizeof(*elf), GFP_KERNEL);
 	if (!elf)
-		goto cleanup;
+		goto out;
 	
 	segs = current->mm->map_count;
 #ifdef ELF_CORE_EXTRA_PHDRS
@@ -2034,8 +2034,9 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file, un
 	set_fs(fs);
 
 cleanup:
-	kfree(elf);
 	free_note_info(&info);
+	kfree(elf);
+out:
 	return has_dumped;
 }
 

fs/compat.c

@@ -1405,7 +1405,7 @@ int compat_do_execve(char * filename,
 		/* execve success */
 		security_bprm_free(bprm);
 		acct_update_integrals(current);
-		kfree(bprm);
+		free_bprm(bprm);
 		return retval;
 	}
 
@@ -1424,7 +1424,7 @@ int compat_do_execve(char * filename,
 	}
 
 out_kfree:
-	kfree(bprm);
+	free_bprm(bprm);
 
 out_ret:
 	return retval;

fs/exec.c

@@ -1251,6 +1251,12 @@ int search_binary_handler(struct linux_binprm *bprm,struct pt_regs *regs)
 
 EXPORT_SYMBOL(search_binary_handler);
 
+void free_bprm(struct linux_binprm *bprm)
+{
+	free_arg_pages(bprm);
+	kfree(bprm);
+}
+
 /*
  * sys_execve() executes a new program.
  */
@@ -1320,17 +1326,15 @@ int do_execve(char * filename,
 	retval = search_binary_handler(bprm,regs);
 	if (retval >= 0) {
 		/* execve success */
-		free_arg_pages(bprm);
 		security_bprm_free(bprm);
 		acct_update_integrals(current);
-		kfree(bprm);
+		free_bprm(bprm);
 		if (displaced)
 			put_files_struct(displaced);
 		return retval;
 	}
 
 out:
-	free_arg_pages(bprm);
 	if (bprm->security)
 		security_bprm_free(bprm);
 
@@ -1344,7 +1348,7 @@ int do_execve(char * filename,
 		fput(bprm->file);
 	}
 out_kfree:
-	kfree(bprm);
+	free_bprm(bprm);
 
 out_files:
 	if (displaced)

fs/file.c

@@ -26,6 +26,8 @@ struct fdtable_defer {
 };
 
 int sysctl_nr_open __read_mostly = 1024*1024;
+int sysctl_nr_open_min = BITS_PER_LONG;
+int sysctl_nr_open_max = 1024 * 1024; /* raised later */
 
 /*
  * We use this list to defer free fdtables that have vmalloced
@@ -119,8 +121,6 @@ static void copy_fdtable(struct fdtable *nfdt, struct fdtable *ofdt)
 	unsigned int cpy, set;
 
 	BUG_ON(nfdt->max_fds < ofdt->max_fds);
-	if (ofdt->max_fds == 0)
-		return;
 
 	cpy = ofdt->max_fds * sizeof(struct file *);
 	set = (nfdt->max_fds - ofdt->max_fds) * sizeof(struct file *);
@@ -261,6 +261,139 @@ int expand_files(struct files_struct *files, int nr)
 	return expand_fdtable(files, nr);
 }
 
+static int count_open_files(struct fdtable *fdt)
+{
+	int size = fdt->max_fds;
+	int i;
+
+	/* Find the last open fd */
+	for (i = size/(8*sizeof(long)); i > 0; ) {
+		if (fdt->open_fds->fds_bits[--i])
+			break;
+	}
+	i = (i+1) * 8 * sizeof(long);
+	return i;
+}
+
+/*
+ * Allocate a new files structure and copy contents from the
+ * passed in files structure.
+ * errorp will be valid only when the returned files_struct is NULL.
+ */
+struct files_struct *dup_fd(struct files_struct *oldf, int *errorp)
+{
+	struct files_struct *newf;
+	struct file **old_fds, **new_fds;
+	int open_files, size, i;
+	struct fdtable *old_fdt, *new_fdt;
+
+	*errorp = -ENOMEM;
+	newf = kmem_cache_alloc(files_cachep, GFP_KERNEL);
+	if (!newf)
+		goto out;
+
+	atomic_set(&newf->count, 1);
+
+	spin_lock_init(&newf->file_lock);
+	newf->next_fd = 0;
+	new_fdt = &newf->fdtab;
+	new_fdt->max_fds = NR_OPEN_DEFAULT;
+	new_fdt->close_on_exec = (fd_set *)&newf->close_on_exec_init;
+	new_fdt->open_fds = (fd_set *)&newf->open_fds_init;
+	new_fdt->fd = &newf->fd_array[0];
+	INIT_RCU_HEAD(&new_fdt->rcu);
+	new_fdt->next = NULL;
+
+	spin_lock(&oldf->file_lock);
+	old_fdt = files_fdtable(oldf);
+	open_files = count_open_files(old_fdt);
+
+	/*
+	 * Check whether we need to allocate a larger fd array and fd set.
+	 */
+	while (unlikely(open_files > new_fdt->max_fds)) {
+		spin_unlock(&oldf->file_lock);
+
+		if (new_fdt != &newf->fdtab) {
+			free_fdarr(new_fdt);
+			free_fdset(new_fdt);
+			kfree(new_fdt);
+		}
+
+		new_fdt = alloc_fdtable(open_files - 1);
+		if (!new_fdt) {
+			*errorp = -ENOMEM;
+			goto out_release;
+		}
+
+		/* beyond sysctl_nr_open; nothing to do */
+		if (unlikely(new_fdt->max_fds < open_files)) {
+			free_fdarr(new_fdt);
+			free_fdset(new_fdt);
+			kfree(new_fdt);
+			*errorp = -EMFILE;
+			goto out_release;
+		}
+
+		/*
+		 * Reacquire the oldf lock and a pointer to its fd table
+		 * who knows it may have a new bigger fd table. We need
+		 * the latest pointer.
+		 */
+		spin_lock(&oldf->file_lock);
+		old_fdt = files_fdtable(oldf);
+		open_files = count_open_files(old_fdt);
+	}
+
+	old_fds = old_fdt->fd;
+	new_fds = new_fdt->fd;
+
+	memcpy(new_fdt->open_fds->fds_bits,
+		old_fdt->open_fds->fds_bits, open_files/8);
+	memcpy(new_fdt->close_on_exec->fds_bits,
+		old_fdt->close_on_exec->fds_bits, open_files/8);
+
+	for (i = open_files; i != 0; i--) {
+		struct file *f = *old_fds++;
+		if (f) {
+			get_file(f);
+		} else {
+			/*
+			 * The fd may be claimed in the fd bitmap but not yet
+			 * instantiated in the files array if a sibling thread
+			 * is partway through open().  So make sure that this
+			 * fd is available to the new process.
+			 */
+			FD_CLR(open_files - i, new_fdt->open_fds);
+		}
+		rcu_assign_pointer(*new_fds++, f);
+	}
+	spin_unlock(&oldf->file_lock);
+
+	/* compute the remainder to be cleared */
+	size = (new_fdt->max_fds - open_files) * sizeof(struct file *);
+
+	/* This is long word aligned thus could use a optimized version */
+	memset(new_fds, 0, size);
+
+	if (new_fdt->max_fds > open_files) {
+		int left = (new_fdt->max_fds-open_files)/8;
+		int start = open_files / (8 * sizeof(unsigned long));
+
+		memset(&new_fdt->open_fds->fds_bits[start], 0, left);
+		memset(&new_fdt->close_on_exec->fds_bits[start], 0, left);
+	}
+
+	rcu_assign_pointer(newf->fdt, new_fdt);
+
+	return newf;
+
+out_release:
+	kmem_cache_free(files_cachep, newf);
+out:
+	return NULL;
+}
+
 static void __devinit fdtable_defer_list_init(int cpu)
 {
 	struct fdtable_defer *fddef = &per_cpu(fdtable_defer_list, cpu);
@@ -274,4 +407,19 @@ void __init files_defer_init(void)
 	int i;
 	for_each_possible_cpu(i)
 		fdtable_defer_list_init(i);
+	sysctl_nr_open_max = min((size_t)INT_MAX, ~(size_t)0/sizeof(void *)) &
+			     -BITS_PER_LONG;
 }
+
+struct files_struct init_files = {
+	.count		= ATOMIC_INIT(1),
+	.fdt		= &init_files.fdtab,
+	.fdtab		= {
+		.max_fds	= NR_OPEN_DEFAULT,
+		.fd		= &init_files.fd_array[0],
+		.close_on_exec	= (fd_set *)&init_files.close_on_exec_init,
+		.open_fds	= (fd_set *)&init_files.open_fds_init,
+		.rcu		= RCU_HEAD_INIT,
+	},
+	.file_lock	= __SPIN_LOCK_UNLOCKED(init_task.file_lock),
+};

fs/namei.c

@@ -2003,18 +2003,22 @@ struct dentry *lookup_create(struct nameidata *nd, int is_dir)
 	if (IS_ERR(dentry))
 		goto fail;
 
+	if (dentry->d_inode)
+		goto eexist;
 	/*
 	 * Special case - lookup gave negative, but... we had foo/bar/
 	 * From the vfs_mknod() POV we just have a negative dentry -
 	 * all is fine. Let's be bastards - you had / on the end, you've
 	 * been asking for (non-existent) directory. -ENOENT for you.
 	 */
-	if (!is_dir && nd->last.name[nd->last.len] && !dentry->d_inode)
+	if (unlikely(!is_dir && nd->last.name[nd->last.len])) {
-		goto enoent;
+		dput(dentry);
+		dentry = ERR_PTR(-ENOENT);
+	}
 	return dentry;
-enoent:
+eexist:
 	dput(dentry);
-	dentry = ERR_PTR(-ENOENT);
+	dentry = ERR_PTR(-EEXIST);
 fail:
 	return dentry;
 }

include/linux/binfmts.h

@@ -99,6 +99,7 @@ extern int copy_strings_kernel(int argc,char ** argv,struct linux_binprm *bprm);
 extern void compute_creds(struct linux_binprm *binprm);
 extern int do_coredump(long signr, int exit_code, struct pt_regs * regs);
 extern int set_binfmt(struct linux_binfmt *new);
+extern void free_bprm(struct linux_binprm *);
 
 #endif /* __KERNEL__ */
 #endif /* _LINUX_BINFMTS_H */

include/linux/fdtable.h

@@ -93,6 +93,7 @@ struct files_struct *get_files_struct(struct task_struct *);
 void put_files_struct(struct files_struct *fs);
 void reset_files_struct(struct files_struct *);
 int unshare_files(struct files_struct **);
+struct files_struct *dup_fd(struct files_struct *, int *);
 
 extern struct kmem_cache *files_cachep;
 

include/linux/init_task.h

@@ -1,7 +1,6 @@
 #ifndef _LINUX__INIT_TASK_H
 #define _LINUX__INIT_TASK_H
 
-#include <linux/fdtable.h>
 #include <linux/rcupdate.h>
 #include <linux/irqflags.h>
 #include <linux/utsname.h>
@@ -12,27 +11,7 @@
 #include <linux/securebits.h>
 #include <net/net_namespace.h>
 
-#define INIT_FDTABLE \
+extern struct files_struct init_files;
-{							\
-	.max_fds	= NR_OPEN_DEFAULT, 		\
-	.fd		= &init_files.fd_array[0], 	\
-	.close_on_exec	= (fd_set *)&init_files.close_on_exec_init, \
-	.open_fds	= (fd_set *)&init_files.open_fds_init, 	\
-	.rcu		= RCU_HEAD_INIT, 		\
-	.next		= NULL,		 		\
-}
-
-#define INIT_FILES \
-{ 							\
-	.count		= ATOMIC_INIT(1), 		\
-	.fdt		= &init_files.fdtab, 		\
-	.fdtab		= INIT_FDTABLE,			\
-	.file_lock	= __SPIN_LOCK_UNLOCKED(init_task.file_lock), \
-	.next_fd	= 0, 				\
-	.close_on_exec_init = { { 0, } }, 		\
-	.open_fds_init	= { { 0, } }, 			\
-	.fd_array	= { NULL, } 			\
-}
 
 #define INIT_KIOCTX(name, which_mm) \
 {							\

kernel/fork.c

@@ -660,136 +660,6 @@ static int copy_fs(unsigned long clone_flags, struct task_struct *tsk)
 	return 0;
 }
 
-static int count_open_files(struct fdtable *fdt)
-{
-	int size = fdt->max_fds;
-	int i;
-
-	/* Find the last open fd */
-	for (i = size/(8*sizeof(long)); i > 0; ) {
-		if (fdt->open_fds->fds_bits[--i])
-			break;
-	}
-	i = (i+1) * 8 * sizeof(long);
-	return i;
-}
-
-static struct files_struct *alloc_files(void)
-{
-	struct files_struct *newf;
-	struct fdtable *fdt;
-
-	newf = kmem_cache_alloc(files_cachep, GFP_KERNEL);
-	if (!newf)
-		goto out;
-
-	atomic_set(&newf->count, 1);
-
-	spin_lock_init(&newf->file_lock);
-	newf->next_fd = 0;
-	fdt = &newf->fdtab;
-	fdt->max_fds = NR_OPEN_DEFAULT;
-	fdt->close_on_exec = (fd_set *)&newf->close_on_exec_init;
-	fdt->open_fds = (fd_set *)&newf->open_fds_init;
-	fdt->fd = &newf->fd_array[0];
-	INIT_RCU_HEAD(&fdt->rcu);
-	fdt->next = NULL;
-	rcu_assign_pointer(newf->fdt, fdt);
-out:
-	return newf;
-}
-
-/*
- * Allocate a new files structure and copy contents from the
- * passed in files structure.
- * errorp will be valid only when the returned files_struct is NULL.
- */
-static struct files_struct *dup_fd(struct files_struct *oldf, int *errorp)
-{
-	struct files_struct *newf;
-	struct file **old_fds, **new_fds;
-	int open_files, size, i;
-	struct fdtable *old_fdt, *new_fdt;
-
-	*errorp = -ENOMEM;
-	newf = alloc_files();
-	if (!newf)
-		goto out;
-
-	spin_lock(&oldf->file_lock);
-	old_fdt = files_fdtable(oldf);
-	new_fdt = files_fdtable(newf);
-	open_files = count_open_files(old_fdt);
-
-	/*
-	 * Check whether we need to allocate a larger fd array and fd set.
-	 * Note: we're not a clone task, so the open count won't change.
-	 */
-	if (open_files > new_fdt->max_fds) {
-		new_fdt->max_fds = 0;
-		spin_unlock(&oldf->file_lock);
-		spin_lock(&newf->file_lock);
-		*errorp = expand_files(newf, open_files-1);
-		spin_unlock(&newf->file_lock);
-		if (*errorp < 0)
-			goto out_release;
-		new_fdt = files_fdtable(newf);
-		/*
-		 * Reacquire the oldf lock and a pointer to its fd table
-		 * who knows it may have a new bigger fd table. We need
-		 * the latest pointer.
-		 */
-		spin_lock(&oldf->file_lock);
-		old_fdt = files_fdtable(oldf);
-	}
-
-	old_fds = old_fdt->fd;
-	new_fds = new_fdt->fd;
-
-	memcpy(new_fdt->open_fds->fds_bits,
-		old_fdt->open_fds->fds_bits, open_files/8);
-	memcpy(new_fdt->close_on_exec->fds_bits,
-		old_fdt->close_on_exec->fds_bits, open_files/8);
-
-	for (i = open_files; i != 0; i--) {
-		struct file *f = *old_fds++;
-		if (f) {
-			get_file(f);
-		} else {
-			/*
-			 * The fd may be claimed in the fd bitmap but not yet
-			 * instantiated in the files array if a sibling thread
-			 * is partway through open().  So make sure that this
-			 * fd is available to the new process.
-			 */
-			FD_CLR(open_files - i, new_fdt->open_fds);
-		}
-		rcu_assign_pointer(*new_fds++, f);
-	}
-	spin_unlock(&oldf->file_lock);
-
-	/* compute the remainder to be cleared */
-	size = (new_fdt->max_fds - open_files) * sizeof(struct file *);
-
-	/* This is long word aligned thus could use a optimized version */
-	memset(new_fds, 0, size);
-
-	if (new_fdt->max_fds > open_files) {
-		int left = (new_fdt->max_fds-open_files)/8;
-		int start = open_files / (8 * sizeof(unsigned long));
-
-		memset(&new_fdt->open_fds->fds_bits[start], 0, left);
-		memset(&new_fdt->close_on_exec->fds_bits[start], 0, left);
-	}
-
-	return newf;
-
-out_release:
-	kmem_cache_free(files_cachep, newf);
-out:
-	return NULL;
-}
-
 static int copy_files(unsigned long clone_flags, struct task_struct * tsk)
 {
 	struct files_struct *oldf, *newf;

kernel/sysctl.c

@@ -81,6 +81,7 @@ extern int compat_log;
 extern int maps_protect;
 extern int sysctl_stat_interval;
 extern int latencytop_enabled;
+extern int sysctl_nr_open_min, sysctl_nr_open_max;
 
 /* Constants used for minimum and  maximum */
 #if defined(CONFIG_DETECT_SOFTLOCKUP) || defined(CONFIG_HIGHMEM)
@@ -1190,7 +1191,9 @@ static struct ctl_table fs_table[] = {
 		.data		= &sysctl_nr_open,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec,
+		.proc_handler	= &proc_dointvec_minmax,
+		.extra1		= &sysctl_nr_open_min,
+		.extra2		= &sysctl_nr_open_max,
 	},
 	{
 		.ctl_name	= FS_DENTRY,