Fix over-zealous flush_disk when changing device size.

There are two cases when we call flush_disk.
In one, the device has disappeared (check_disk_change) so any
data will hold becomes irrelevant.
In the oter, the device has changed size (check_disk_size_change)
so data we hold may be irrelevant.

In both cases it makes sense to discard any 'clean' buffers,
so they will be read back from the device if needed.

In the former case it makes sense to discard 'dirty' buffers
as there will never be anywhere safe to write the data.  In the
second case it *does*not* make sense to discard dirty buffers
as that will lead to file system corruption when you simply enlarge
the containing devices.

flush_disk calls __invalidate_devices.
__invalidate_device calls both invalidate_inodes and invalidate_bdev.

invalidate_inodes *does* discard I_DIRTY inodes and this does lead
to fs corruption.

invalidate_bev *does*not* discard dirty pages, but I don't really care
about that at present.

So this patch adds a flag to __invalidate_device (calling it
__invalidate_device2) to indicate whether dirty buffers should be
killed, and this is passed to invalidate_inodes which can choose to
skip dirty inodes.

flusk_disk then passes true from check_disk_change and false from
check_disk_size_change.

dm avoids tripping over this problem by calling i_size_write directly
rathher than using check_disk_size_change.

md does use check_disk_size_change and so is affected.

This regression was introduced by commit 608aeef17a which causes
check_disk_size_change to call flush_disk, so it is suitable for any
kernel since 2.6.27.

Cc: stable@kernel.org
Acked-by: Jeff Moyer <jmoyer@redhat.com>
Cc: Andrew Patterson <andrew.patterson@hp.com>
Cc: Jens Axboe <axboe@kernel.dk>
Signed-off-by: NeilBrown <neilb@suse.de>

diff --git a/block/genhd.c b/block/genhd.c
index 6a5b772aa2016f52f986a2cca396daf94ad0c7c4..cbf1112a885c0c715e4d38a3dfad91f74cd1f1fa 100644 (file)
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -1355,7 +1355,7 @@ int invalidate_partition(struct gendisk *disk, int partno)
        struct block_device *bdev = bdget_disk(disk, partno);
        if (bdev) {
                fsync_bdev(bdev);
-               res = __invalidate_device(bdev);
+               res = __invalidate_device(bdev, true);
                bdput(bdev);
        }
        return res;

diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c
index b9ba04fc2b34eb0845cf0c724b64151e53b05c65..77fc76f8aea91b79be3f8cfc6c640e3e23748de0 100644 (file)
--- a/drivers/block/floppy.c
+++ b/drivers/block/floppy.c
@@ -3281,7 +3281,7 @@ static int set_geometry(unsigned int cmd, struct floppy_struct *g,
                        struct block_device *bdev = opened_bdev[cnt];
                        if (!bdev || ITYPE(drive_state[cnt].fd_device) != type)
                                continue;
-                       __invalidate_device(bdev);
+                       __invalidate_device(bdev, true);
                }
                mutex_unlock(&open_lock);
        } else {

diff --git a/fs/block_dev.c b/fs/block_dev.c
index 333a7bb4cb9c0865543bc210c6df2613662950c9..5e23152d04ad02b8c871beb230e95389a5c79c90 100644 (file)
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -927,9 +927,9 @@ EXPORT_SYMBOL_GPL(bd_unlink_disk_holder);
  * when a disk has been changed -- either by a media change or online
  * resize.
  */
-static void flush_disk(struct block_device *bdev)
+static void flush_disk(struct block_device *bdev, bool kill_dirty)
 {
-       if (__invalidate_device(bdev)) {
+       if (__invalidate_device(bdev, kill_dirty)) {
                char name[BDEVNAME_SIZE] = "";
 
                if (bdev->bd_disk)
@@ -966,7 +966,7 @@ void check_disk_size_change(struct gendisk *disk, struct block_device *bdev)
                       "%s: detected capacity change from %lld to %lld\n",
                       name, bdev_size, disk_size);
                i_size_write(bdev->bd_inode, disk_size);
-               flush_disk(bdev);
+               flush_disk(bdev, false);
        }
 }
 EXPORT_SYMBOL(check_disk_size_change);
@@ -1019,7 +1019,7 @@ int check_disk_change(struct block_device *bdev)
        if (!(events & DISK_EVENT_MEDIA_CHANGE))
                return 0;
 
-       flush_disk(bdev);
+       flush_disk(bdev, true);
        if (bdops->revalidate_disk)
                bdops->revalidate_disk(bdev->bd_disk);
        return 1;
@@ -1601,7 +1601,7 @@ fail:
 }
 EXPORT_SYMBOL(lookup_bdev);
 
-int __invalidate_device(struct block_device *bdev)
+int __invalidate_device(struct block_device *bdev, bool kill_dirty)
 {
        struct super_block *sb = get_super(bdev);
        int res = 0;
@@ -1614,7 +1614,7 @@ int __invalidate_device(struct block_device *bdev)
                 * hold).
                 */
                shrink_dcache_sb(sb);
-               res = invalidate_inodes(sb);
+               res = invalidate_inodes(sb, kill_dirty);
                drop_super(sb);
        }
        invalidate_bdev(bdev);

diff --git a/fs/inode.c b/fs/inode.c
index da85e56378f3e1d98c241901171e24e83882f189..c50d7feb87b1203b62df1f15d065d1dc109b6120 100644 (file)
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -540,11 +540,14 @@ void evict_inodes(struct super_block *sb)
 /**
  * invalidate_inodes   - attempt to free all inodes on a superblock
  * @sb:                superblock to operate on
+ * @kill_dirty: flag to guide handling of dirty inodes
  *
  * Attempts to free all inodes for a given superblock.  If there were any
  * busy inodes return a non-zero value, else zero.
+ * If @kill_dirty is set, discard dirty inodes too, otherwise treat
+ * them as busy.
  */
-int invalidate_inodes(struct super_block *sb)
+int invalidate_inodes(struct super_block *sb, bool kill_dirty)
 {
        int busy = 0;
        struct inode *inode, *next;
@@ -556,6 +559,10 @@ int invalidate_inodes(struct super_block *sb)
        list_for_each_entry_safe(inode, next, &sb->s_inodes, i_sb_list) {
                if (inode->i_state & (I_NEW | I_FREEING | I_WILL_FREE))
                        continue;
+               if (inode->i_state & I_DIRTY && !kill_dirty) {
+                       busy = 1;
+                       continue;
+               }
                if (atomic_read(&inode->i_count)) {
                        busy = 1;
                        continue;

diff --git a/fs/internal.h b/fs/internal.h
index 0663568b1247f88a0ca57ad19b8f9c7d36680f8e..9b976b57d7fe3b4b731cbfe3d90b4918c0c8b730 100644 (file)
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -112,4 +112,4 @@ extern void release_open_intent(struct nameidata *);
  */
 extern int get_nr_dirty_inodes(void);
 extern void evict_inodes(struct super_block *);
-extern int invalidate_inodes(struct super_block *);
+extern int invalidate_inodes(struct super_block *, bool);

diff --git a/include/linux/fs.h b/include/linux/fs.h
index 32b38cd829d33561163dd340ce1e307150db25d3..683f4c566c822880119a888958f61fe7bb1c0881 100644 (file)
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -2139,7 +2139,7 @@ extern void check_disk_size_change(struct gendisk *disk,
                                   struct block_device *bdev);
 extern int revalidate_disk(struct gendisk *);
 extern int check_disk_change(struct block_device *);
-extern int __invalidate_device(struct block_device *);
+extern int __invalidate_device(struct block_device *, bool);
 extern int invalidate_partition(struct gendisk *, int);
 #endif
 unsigned long invalidate_mapping_pages(struct address_space *mapping,