Am 19.06.2013 16:02, schrieb j.glisse at gmail.com:
> From: Jerome Glisse <jglisse at redhat.com>
>
> There might be issue with lockup detection when scheduling on an
> empty ring that have been sitting idle for a while. Thus update
> the lockup tracking data when scheduling new work in an empty ring.
>
> Signed-off-by: Jerome Glisse <jglisse at redhat.com>
> Tested-by: Andy Lutomirski <luto at amacapital.net>
> Cc: stable at vger.kernel.org
Reviewed-by: Christian K?nig <christian.koenig at amd.com>
> ---
> drivers/gpu/drm/radeon/radeon_ring.c | 7 +++++++
> 1 file changed, 7 insertions(+)
>
> diff --git a/drivers/gpu/drm/radeon/radeon_ring.c
> b/drivers/gpu/drm/radeon/radeon_ring.c
> index e17faa7..82434018 100644
> --- a/drivers/gpu/drm/radeon/radeon_ring.c
> +++ b/drivers/gpu/drm/radeon/radeon_ring.c
> @@ -402,6 +402,13 @@ int radeon_ring_alloc(struct radeon_device *rdev, struct
> radeon_ring *ring, unsi
> return -ENOMEM;
> /* Align requested size with padding so unlock_commit can
> * pad safely */
> + radeon_ring_free_size(rdev, ring);
> + if (ring->ring_free_dw == (ring->ring_size / 4)) {
> + /* This is an empty ring update lockup info to avoid
> + * false positive.
> + */
> + radeon_ring_lockup_update(ring);
> + }
> ndw = (ndw + ring->align_mask) & ~ring->align_mask;
> while (ndw > (ring->ring_free_dw - 1)) {
> radeon_ring_free_size(rdev, ring);
