Linux 2.6.37 compat, WRITE_FLUSH_FUA

The WRITE_FLUSH, WRITE_FUA, and WRITE_FLUSH_FUA flags have been
introduced as a replacement for WRITE_BARRIER.  This was done
to allow richer semantics to be expressed to the block layer.
It is the block layers responsibility to choose the correct way
to implement these semantics.

This change simply updates the bio's to use the new kernel API
which should be absolutely safe.  However, since ZFS depends
entirely on this working as designed for correctness we do
want to be careful.

Closes #281

diff --git a/include/linux/blkdev_compat.h b/include/linux/blkdev_compat.h
index 3bf1f45..0a7c985 100644 (file)
--- a/include/linux/blkdev_compat.h
+++ b/include/linux/blkdev_compat.h
@@ -339,6 +339,20 @@ bio_set_flags_failfast(struct block_device *bdev, int *flags)
 #endif
 
 /*
+ * 2.6.37 API change
+ * The WRITE_FLUSH, WRITE_FUA, and WRITE_FLUSH_FUA flags have been
+ * introduced as a replacement for WRITE_BARRIER.  This was done to
+ * allow richer semantics to be expressed to the block layer.  It is
+ * the block layers responsibility to choose the correct way to
+ * implement these semantics.
+ */
+#ifdef WRITE_FLUSH_FUA
+# define VDEV_WRITE_FLUSH_FUA          WRITE_FLUSH_FUA
+#else
+# define VDEV_WRITE_FLUSH_FUA          WRITE_BARRIER
+#endif
+
+/*
  * Default Linux IO Scheduler,
  * Setting the scheduler to noop will allow the Linux IO scheduler to
  * still perform front and back merging, while leaving the request

diff --git a/module/zfs/vdev_disk.c b/module/zfs/vdev_disk.c
index 4145dec..28a4861 100644 (file)
--- a/module/zfs/vdev_disk.c
+++ b/module/zfs/vdev_disk.c
@@ -559,7 +559,7 @@ vdev_disk_io_flush(struct block_device *bdev, zio_t *zio)
        bio->bi_private = zio;
        bio->bi_bdev = bdev;
        zio->io_delay = jiffies_64;
-       submit_bio(WRITE_BARRIER, bio);
+       submit_bio(VDEV_WRITE_FLUSH_FUA, bio);
 
        return 0;
 }