On Fri, Aug 10, 2012 at 1:29 PM, <alexdeucher@xxxxxxxxx> wrote:
> From: Alex Deucher <alexander.deucher@xxxxxxx>
>
> 1. Drop gui idle stuff, it's not as reliable as fences and only
> covers the 3D engine.
> 2. Wait for fences on all rings. This makes sure all rings are
> idle when reclocking.
>
> Signed-off-by: Alex Deucher <alexander.deucher@xxxxxxx>
Reviewed-by: Jerome Glisse <jglisse@xxxxxxxxxx>
> ---
> drivers/gpu/drm/radeon/radeon_pm.c | 17 ++++++-----------
> 1 files changed, 6 insertions(+), 11 deletions(-)
>
> diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
> index 7ae6066..2c2c901 100644
> --- a/drivers/gpu/drm/radeon/radeon_pm.c
> +++ b/drivers/gpu/drm/radeon/radeon_pm.c
> @@ -253,18 +253,13 @@ static void radeon_pm_set_clocks(struct radeon_device *rdev)
> down_write(&rdev->pm.mclk_lock);
> mutex_lock(&rdev->ring_lock);
>
> - /* gui idle int has issues on older chips it seems */
> - if (rdev->family >= CHIP_R600) {
> - if (rdev->irq.installed) {
> - /* wait for GPU to become idle */
> - radeon_irq_kms_wait_gui_idle(rdev);
> - }
> - } else {
> - struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
> - if (ring->ready) {
> - radeon_fence_wait_empty_locked(rdev, RADEON_RING_TYPE_GFX_INDEX);
> - }
> + /* wait for the rings to drain */
> + for (i = 0; i < RADEON_NUM_RINGS; i++) {
> + struct radeon_ring *ring = &rdev->ring[i];
> + if (ring->ready)
> + radeon_fence_wait_empty_locked(rdev, i);
> }
> +
> radeon_unmap_vram_bos(rdev);
>
> if (rdev->irq.installed) {
> --
> 1.7.7.5
>
> _______________________________________________
> dri-devel mailing list
> dri-devel@xxxxxxxxxxxxxxxxxxxxx
> http://lists.freedesktop.org/mailman/listinfo/dri-devel
_______________________________________________
dri-devel mailing list
dri-devel@xxxxxxxxxxxxxxxxxxxxx
http://lists.freedesktop.org/mailman/listinfo/dri-devel
