Use f_lock to protect f_flags

Traditionally, changes to struct file->f_flags have been done under BKL
protection, or with no protection at all.  This patch causes all f_flags
changes after file open/creation time to be done under protection of
f_lock.  This allows the removal of some BKL usage and fixes a number of
longstanding (if microscopic) races.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Al Viro <viro@ZenIV.linux.org.uk>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>

3 files changed, 10 insertions, 4 deletions

diff --git a/fs/fcntl.c b/fs/fcntl.c
index bd215cc791da..04df8570a2d2 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -189,7 +189,9 @@ static int setfl(int fd, struct file * filp, unsigned long arg)
                 }
         }
 
+        spin_lock(&filp->f_lock);
         filp->f_flags = (arg & SETFL_MASK) | (filp->f_flags & ~SETFL_MASK);
+        spin_unlock(&filp->f_lock);
  out:
         unlock_kernel();
         return error;
diff --git a/fs/ioctl.c b/fs/ioctl.c
index 240ec63984cb..421aab465dab 100644
--- a/fs/ioctl.c
+++ b/fs/ioctl.c
@@ -404,10 +404,12 @@ static int ioctl_fionbio(struct file *filp, int __user *argp)
         if (O_NONBLOCK != O_NDELAY)
                 flag |= O_NDELAY;
 #endif
+        spin_lock(&filp->f_lock);
         if (on)
                 filp->f_flags |= flag;
         else
                 filp->f_flags &= ~flag;
+        spin_unlock(&filp->f_lock);
         return error;
 }
 
@@ -432,10 +434,12 @@ static int ioctl_fioasync(unsigned int fd, struct file *filp,
         if (error)
                 return error;
 
+        spin_lock(&filp->f_lock);
         if (on)
                 filp->f_flags |= FASYNC;
         else
                 filp->f_flags &= ~FASYNC;
+        spin_unlock(&filp->f_lock);
         return error;
 }
 
@@ -499,10 +503,7 @@ int do_vfs_ioctl(struct file *filp, unsigned int fd, unsigned int cmd,
                 break;
 
         case FIONBIO:
-                /* BKL needed to avoid races tweaking f_flags */
-                lock_kernel();
                 error = ioctl_fionbio(filp, argp);
-                unlock_kernel();
                 break;
 
         case FIOASYNC:
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 6e50aaa56ca2..c165a6403df0 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -998,8 +998,11 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file,
 
         if (!EX_ISSYNC(exp))
                 stable = 0;
-        if (stable && !EX_WGATHER(exp))
+        if (stable && !EX_WGATHER(exp)) {
+                spin_lock(&file->f_lock);
                 file->f_flags |= O_SYNC;
+                spin_unlock(&file->f_lock);
+        }
 
         /* Write the data. */
         oldfs = get_fs(); set_fs(KERNEL_DS);