Commit 9bdd9bd6 authored May 18, 2016 by Brian Foster Committed by Dave Chinner May 18, 2016

xfs: buffer ->bi_end_io function requires irq-safe lock



Reports have surfaced of a lockdep splat complaining about an
irq-safe -> irq-unsafe locking order in the xfs_buf_bio_end_io() bio
completion handler. This only occurs when I/O errors are present
because bp->b_lock is only acquired in this context to protect
setting an error on the buffer. The problem is that this lock can be
acquired with the (request_queue) q->queue_lock held. See
scsi_end_request() or ata_qc_schedule_eh(), for example.

Replace the locked test/set of b_io_error with a cmpxchg() call.
This eliminates the need for the lock and thus the lock ordering
problem goes away.

Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>

parent f55532a0

fs/xfs/xfs_buf.c

+4 −8

Original line number	Diff line number	Diff line
		@@ -1100,22 +1100,18 @@ xfs_bwrite(
		return error;
		}

		STATIC void
		static void
		xfs_buf_bio_end_io(
		struct bio *bio)
		{
		xfs_buf_t bp = (xfs_buf_t )bio->bi_private;
		struct xfs_buf bp = (struct xfs_buf )bio->bi_private;

		/*
		* don't overwrite existing errors - otherwise we can lose errors on
		* buffers that require multiple bios to complete.
		*/
		if (bio->bi_error) {
		spin_lock(&bp->b_lock);
		if (!bp->b_io_error)
		bp->b_io_error = bio->bi_error;
		spin_unlock(&bp->b_lock);
		}
		if (bio->bi_error)
		cmpxchg(&bp->b_io_error, 0, bio->bi_error);

		if (!bp->b_error && xfs_buf_is_vmapped(bp) && (bp->b_flags & XBF_READ))
		invalidate_kernel_vmap_range(bp->b_addr, xfs_buf_vmap_len(bp));