Re: [RFC][PATCH v2] readahead: introduce O_RANDOM_READ for POSIX_FADV_RANDOM

[Wu Fengguang <fengguang.wu@intel.com>]

On Fri, Jan 08, 2010 at 09:08:28PM +0800, Christoph Hellwig wrote:
> Just thinking about this again, why don't you put the flag into
> file->f_mode and the FMODE_* namespace given that we don't want it to be
> settable from open?

Good idea. To do that without race I would like to add ->f_lock to 
f_mode modifications at non-open time, like this. What do you think?

---
vfs: take f_lock on modifying f_mode after open time

Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
---
 fs/file_table.c     |    2 ++
 fs/nfsd/nfs4state.c |    2 ++
 2 files changed, 4 insertions(+)

--- linux.orig/fs/file_table.c	2010-01-09 10:50:24.000000000 +0800
+++ linux/fs/file_table.c	2010-01-09 10:51:06.000000000 +0800
@@ -394,7 +394,9 @@ retry:
 			continue;
 		if (!(f->f_mode & FMODE_WRITE))
 			continue;
+		spin_lock(&f->f_lock);
 		f->f_mode &= ~FMODE_WRITE;
+		spin_unlock(&f->f_lock);
 		if (file_check_writeable(f) != 0)
 			continue;
 		file_release_write(f);
--- linux.orig/fs/nfsd/nfs4state.c	2010-01-09 10:52:54.000000000 +0800
+++ linux/fs/nfsd/nfs4state.c	2010-01-09 10:52:56.000000000 +0800
@@ -1998,7 +1998,9 @@ nfs4_file_downgrade(struct file *filp, u
 {
 	if (share_access & NFS4_SHARE_ACCESS_WRITE) {
 		drop_file_write_access(filp);
+		spin_lock(&filp->f_lock);
 		filp->f_mode = (filp->f_mode | FMODE_READ) & ~FMODE_WRITE;
+		spin_unlock(&filp->f_lock);
 	}
 }