The recent fix to the raw driver wasn't quite right: it rewrites the
character-special inode's i_mapping to point back to itself on each close.
So any other currently-open handles against /dev/raw/rawN get a nasty
surprise.
Change it to only rewrite i_mapping on the final close.
Also, change it so that it only redirects its i_mapping on the initial open.
This is not necessary, but is neater.
drivers/char/raw.c | 20 ++++++++++++--------
1 files changed, 12 insertions(+), 8 deletions(-)
diff -puN drivers/char/raw.c~raw-fix-address_space-rewriting drivers/char/raw.c
--- 25/drivers/char/raw.c~raw-fix-address_space-rewriting 2003-03-18 22:02:00.000000000 -0800
+++ 25-akpm/drivers/char/raw.c 2003-03-18 22:09:29.000000000 -0800
@@ -70,10 +70,10 @@ static int raw_open(struct inode *inode,
} else {
err = set_blocksize(bdev, bdev_hardsect_size(bdev));
if (err == 0) {
- raw_devices[minor].inuse++;
- filp->f_dentry->d_inode->i_mapping =
- bdev->bd_inode->i_mapping;
filp->f_flags |= O_DIRECT;
+ if (++raw_devices[minor].inuse == 1)
+ filp->f_dentry->d_inode->i_mapping =
+ bdev->bd_inode->i_mapping;
}
}
}
@@ -83,6 +83,10 @@ out:
return err;
}
+/*
+ * When the final fd which refers to this character-special node is closed, we
+ * make its ->mapping point back at its own i_data.
+ */
static int raw_release(struct inode *inode, struct file *filp)
{
const int minor= minor(inode->i_rdev);
@@ -90,12 +94,12 @@ static int raw_release(struct inode *ino
down(&raw_mutex);
bdev = raw_devices[minor].binding;
- raw_devices[minor].inuse--;
+ if (--raw_devices[minor].inuse == 0) {
+ /* Here inode->i_mapping == bdev->bd_inode->i_mapping */
+ inode->i_mapping = &inode->i_data;
+ inode->i_mapping->backing_dev_info = &default_backing_dev_info;
+ }
up(&raw_mutex);
-
- /* Here inode->i_mapping == bdev->bd_inode->i_mapping */
- inode->i_mapping = &inode->i_data;
- inode->i_mapping->backing_dev_info = &default_backing_dev_info;
bd_release(bdev);
blkdev_put(bdev, BDEV_RAW);
_