New locking/refcounting for fs_struct

* all changes of current->fs are done under task_lock and write_lock of
  old fs->lock
* refcount is not atomic anymore (same protection)
* its decrements are done when removing reference from current; at the
  same time we decide whether to free it.
* put_fs_struct() is gone
* new field - ->in_exec.  Set by check_unsafe_exec() if we are trying to do
  execve() and only subthreads share fs_struct.  Cleared when finishing exec
  (success and failure alike).  Makes CLONE_FS fail with -EAGAIN if set.
* check_unsafe_exec() may fail with -EAGAIN if another execve() from subthread
  is in progress.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

1 files changed, 14 insertions, 2 deletions

diff --git a/fs/compat.c b/fs/compat.c
index 55efdfebdf5a..baabf203b847 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -51,6 +51,7 @@
 #include <linux/poll.h>
 #include <linux/mm.h>
 #include <linux/eventpoll.h>
+#include <linux/fs_struct.h>
 
 #include <asm/uaccess.h>
 #include <asm/mmu_context.h>
@@ -1441,12 +1442,15 @@ int compat_do_execve(char * filename,
         bprm->cred = prepare_exec_creds();
         if (!bprm->cred)
                 goto out_unlock;
-        check_unsafe_exec(bprm);
+
+        retval = check_unsafe_exec(bprm);
+        if (retval)
+                goto out_unlock;
 
         file = open_exec(filename);
         retval = PTR_ERR(file);
         if (IS_ERR(file))
-                goto out_unlock;
+                goto out_unmark;
 
         sched_exec();
 
@@ -1488,6 +1492,9 @@ int compat_do_execve(char * filename,
                 goto out;
 
         /* execve succeeded */
+        write_lock(&current->fs->lock);
+        current->fs->in_exec = 0;
+        write_unlock(&current->fs->lock);
         current->in_execve = 0;
         mutex_unlock(&current->cred_exec_mutex);
         acct_update_integrals(current);
@@ -1506,6 +1513,11 @@ out_file:
                 fput(bprm->file);
         }
 
+out_unmark:
+        write_lock(&current->fs->lock);
+        current->fs->in_exec = 0;
+        write_unlock(&current->fs->lock);
+
 out_unlock:
         current->in_execve = 0;
         mutex_unlock(&current->cred_exec_mutex);