On Fri, Nov 30, 2018 at 05:00:02PM -0500, Sean Paul wrote:
> From: Sean Paul <seanpaul@xxxxxxxxxxxx>
>
> This patch wraps dpu_core_perf_crtc_release_bw() with modeset locks
> since it digs into the state objects.
>
> Changes in v2:
> - None
> Changes in v3:
> - Use those nifty new DRM_MODESET_LOCK_ALL_* helpers (Daniel)
>
> Cc: Daniel Vetter <daniel@xxxxxxxx>
> Cc: Jeykumar Sankaran <jsanka@xxxxxxxxxxxxxx>
> Signed-off-by: Sean Paul <seanpaul@xxxxxxxxxxxx>
> ---
> drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c | 15 ++++++++++++++-
> 1 file changed, 14 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
> index 74ef384d9cd6a..03ddd281a354f 100644
> --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
> +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
> @@ -306,6 +306,19 @@ static void dpu_crtc_vblank_cb(void *data)
> trace_dpu_crtc_vblank_cb(DRMID(crtc));
> }
>
> +static void dpu_crtc_release_bw_unlocked(struct drm_crtc *crtc)
> +{
> + int ret = 0;
> + struct drm_modeset_acquire_ctx ctx;
> +
> + DRM_MODESET_LOCK_ALL_BEGIN(crtc->dev, ctx, 0, ret);
> + dpu_core_perf_crtc_release_bw(crtc);
> + DRM_MODESET_LOCK_ALL_END(ctx, ret);
Sooooo pretty, and correct even :-)
Acked-by: Daniel Vetter <daniel.vetter@xxxxxxxx>
> + if (ret)
> + DRM_ERROR("Failed to acquire modeset locks to release bw, %d\n",
> + ret);
> +}
> +
> static void dpu_crtc_frame_event_work(struct kthread_work *work)
> {
> struct dpu_crtc_frame_event *fevent = container_of(work,
> @@ -335,7 +348,7 @@ static void dpu_crtc_frame_event_work(struct kthread_work *work)
> /* release bandwidth and other resources */
> trace_dpu_crtc_frame_event_done(DRMID(crtc),
> fevent->event);
> - dpu_core_perf_crtc_release_bw(crtc);
> + dpu_crtc_release_bw_unlocked(crtc);
> } else {
> trace_dpu_crtc_frame_event_more_pending(DRMID(crtc),
> fevent->event);
> --
> Sean Paul, Software Engineer, Google / Chromium OS
>
--
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
