Re: [PATCH v3 1/4] drm/vblank: warn on sending stale event

[Daniel Vetter]

On Tue, Jun 25, 2019 at 06:59:12PM +0100, Robert Beckett wrote:
> Warn when about to send stale vblank info and add advice to
> documentation on how to avoid.
> 
> Signed-off-by: Robert Beckett <bob.beckett@xxxxxxxxxxxxx>
> ---
>  drivers/gpu/drm/drm_vblank.c | 17 +++++++++++++++++
>  1 file changed, 17 insertions(+)
> 
> diff --git a/drivers/gpu/drm/drm_vblank.c b/drivers/gpu/drm/drm_vblank.c
> index 603ab105125d..7dabb2bdb733 100644
> --- a/drivers/gpu/drm/drm_vblank.c
> +++ b/drivers/gpu/drm/drm_vblank.c
> @@ -918,6 +918,19 @@ EXPORT_SYMBOL(drm_crtc_arm_vblank_event);
>   *
>   * See drm_crtc_arm_vblank_event() for a helper which can be used in certain
>   * situation, especially to send out events for atomic commit operations.
> + *
> + * Care should be taken to avoid stale timestamps. If:
> + *   - your driver has vblank support (i.e. dev->num_crtcs > 0)
> + *   - the vblank irq is off (i.e. no one called drm_crtc_vblank_get)

drm_crtc_vblank_get() so it becomes a neat hyperlink.

> + *   - from the vblank code's pov the pipe is still running (i.e. not
> + *     in-between a drm_crtc_vblank_off()/on() pair)

Not sure the above will lead to great markup, maybe spell out the
drm_crtc_vblank_on() and maybe make it a bit clearer like "i.e. not after
the call to drm_crtc_vblank_off() but before the next call to
drm_crtc_vblank_on()" so it's clear which said of this pair we're talking
about.

> + * If all of these conditions hold then drm_crtc_send_vblank_event is

style nit: the enumeration is one sentence (and and oxford comman missing
but whatever), but you don't continue it here. Also, does the enumeration
look pretty in the htmldocs output?

> + * going to give you a garbage timestamp and and sequence number (the last
> + * recorded before the irq was disabled). If you call drm_crtc_vblank_get/put
> + * around it, or after vblank_off, then either of those will have rolled things
> + * forward for you.

Again pls fix the markup so all the function reference stick out and work.

> + * So, drivers should call drm_crtc_vblank_off() before this function in their
> + * crtc atomic_disable handlers.

Imo this sentence here is needed but a bit confusing, and we have it
documented already in the atomic_disable hook.

Other option would be to list all the places where a driver might want to
call this and how they could get it wrong, which imo doesn't make sense.

With the nits addressed:

Reviewed-by: Daniel Vetter <daniel.vetter@xxxxxxxx>

>   */
>  void drm_crtc_send_vblank_event(struct drm_crtc *crtc,
>  				struct drm_pending_vblank_event *e)
> @@ -925,8 +938,12 @@ void drm_crtc_send_vblank_event(struct drm_crtc *crtc,
>  	struct drm_device *dev = crtc->dev;
>  	u64 seq;
>  	unsigned int pipe = drm_crtc_index(crtc);
> +	struct drm_vblank_crtc *vblank = &dev->vblank[pipe];
>  	ktime_t now;
>  
> +	WARN_ONCE(dev->num_crtcs > 0 && !vblank->enabled && !vblank->inmodeset,
> +		  "sending stale vblank info\n");
> +
>  	if (dev->num_crtcs > 0) {
>  		seq = drm_vblank_count_and_time(dev, pipe, &now);
>  	} else {
> -- 
> 2.18.0
> 

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch