pipe: set lower and upper limit on max pages in the pipe page array

We need at least two to guarantee proper POSIX behaviour, so
never allow a smaller limit than that.

Also expose a /proc/sys/fs/pipe-max-pages sysctl file that allows
root to define a sane upper limit. Make it default to 16 times the
default size, which is 16 pages.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

fs/pipe.c

@@ -19,10 +19,17 @@
 #include <linux/pagemap.h>
 #include <linux/audit.h>
 #include <linux/syscalls.h>
+#include <linux/fcntl.h>
 
 #include <asm/uaccess.h>
 #include <asm/ioctls.h>
 
+/*
+ * The max size that a non-root user is allowed to grow the pipe. Can
+ * be set by root in /proc/sys/fs/pipe-max-pages
+ */
+unsigned int pipe_max_pages = PIPE_DEF_BUFFERS * 16;
+
 /*
  * We use a start+len construction, which provides full use of the 
  * allocated memory.
@@ -1162,6 +1169,14 @@ long pipe_fcntl(struct file *file, unsigned int cmd, unsigned long arg)
 
 	switch (cmd) {
 	case F_SETPIPE_SZ:
+		if (!capable(CAP_SYS_ADMIN) && arg > pipe_max_pages)
+			return -EINVAL;
+		/*
+		 * The pipe needs to be at least 2 pages large to
+		 * guarantee POSIX behaviour.
+		 */
+		if (arg < 2)
+			return -EINVAL;
 		ret = pipe_set_size(pipe, arg);
 		break;
 	case F_GETPIPE_SZ:

include/linux/pipe_fs_i.h

@@ -139,6 +139,8 @@ void pipe_lock(struct pipe_inode_info *);
 void pipe_unlock(struct pipe_inode_info *);
 void pipe_double_lock(struct pipe_inode_info *, struct pipe_inode_info *);
 
+extern unsigned int pipe_max_pages;
+
 /* Drop the inode semaphore and wait for a pipe event, atomically */
 void pipe_wait(struct pipe_inode_info *pipe);
 

kernel/sysctl.c

@@ -52,6 +52,7 @@
 #include <linux/slow-work.h>
 #include <linux/perf_event.h>
 #include <linux/kprobes.h>
+#include <linux/pipe_fs_i.h>
 
 #include <asm/uaccess.h>
 #include <asm/processor.h>
@@ -1423,6 +1424,14 @@ static struct ctl_table fs_table[] = {
 		.child		= binfmt_misc_table,
 	},
 #endif
+	{
+		.procname	= "pipe-max-pages",
+		.data		= &pipe_max_pages,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec_minmax,
+		.extra1		= &two,
+	},
 /*
  * NOTE: do not add new entries to this table unless you have read
  * Documentation/sysctl/ctl_unnumbered.txt