All of lore.kernel.org
 help / color / mirror / Atom feed
From: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
To: Marcin Slusarz <marcin.slusarz@gmail.com>
Cc: dri-devel@lists.freedesktop.org
Subject: Re: [PATCH] drm: Factor-out drm_emit_vblank_event code. (v2)
Date: Mon, 02 May 2011 09:55:28 +1000	[thread overview]
Message-ID: <1304294128.25088.7.camel@Ed> (raw)
In-Reply-To: <20110429155528.GA3022@joi.lan>


[-- Attachment #1.1: Type: text/plain, Size: 4469 bytes --]

On Fri, 2011-04-29 at 17:55 +0200, Marcin Slusarz wrote:
> On Fri, Apr 29, 2011 at 01:57:10PM +1000, christopher.halse.rogers@canonical.com wrote:
> > From: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
> > 
> > v2: Also pull out the drm_vblank_put call.
> > Signed-off-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
> > ---
> >  drivers/gpu/drm/drm_irq.c |   44 ++++++++++++++++++--------------------------
> >  1 files changed, 18 insertions(+), 26 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
> > index 982ca8c..da56685 100644
> > --- a/drivers/gpu/drm/drm_irq.c
> > +++ b/drivers/gpu/drm/drm_irq.c
> > @@ -931,6 +931,20 @@ void drm_vblank_put(struct drm_device *dev, int crtc)
> >  }
> >  EXPORT_SYMBOL(drm_vblank_put);
> >  
> > +static void drm_emit_vblank_event (struct drm_device *dev,
> > +				   struct drm_pending_vblank_event *e,
> > +				   unsigned int seq, struct timeval *now)
> > +{
> > +	e->event.sequence = seq;
> > +	e->event.tv_sec = now->tv_sec;
> > +	e->event.tv_usec = now->tv_usec;
> > +	drm_vblank_put(dev, e->pipe);
> > +	list_move_tail(&e->base.link, &e->base.file_priv->event_list);
> > +	wake_up_interruptible(&e->base.file_priv->event_wait);
> > +	trace_drm_vblank_event_delivered(e->base.pid, e->pipe,
> > +					 e->event.sequence);
> > +}
> > +
> >  void drm_vblank_off(struct drm_device *dev, int crtc)
> >  {
> >  	struct drm_pending_vblank_event *e, *t;
> > @@ -951,14 +965,7 @@ void drm_vblank_off(struct drm_device *dev, int crtc)
> >  			  wanted %d, current %d\n",
> >  			  e->event.sequence, seq);
> >  
> > -		e->event.sequence = seq;
> > -		e->event.tv_sec = now.tv_sec;
> > -		e->event.tv_usec = now.tv_usec;
> > -		drm_vblank_put(dev, e->pipe);
> > -		list_move_tail(&e->base.link, &e->base.file_priv->event_list);
> > -		wake_up_interruptible(&e->base.file_priv->event_wait);
> > -		trace_drm_vblank_event_delivered(e->base.pid, e->pipe,
> > -						 e->event.sequence);
> > +		drm_emit_vblank_event(dev, e, seq, &now);
> >  	}
> >  
> >  	WARN_ON(atomic_read(&dev->vblank_refcount[crtc]) != 0);
> > @@ -1104,18 +1111,10 @@ static int drm_queue_vblank_event(struct drm_device *dev, int pipe,
> >  				      vblwait->request.sequence);
> >  
> >  	e->event.sequence = vblwait->request.sequence;
> > +	list_add_tail(&e->base.link, &dev->vblank_event_list);
> 
> Is &dev->vblank_event_list == &e->base.file_priv->event_list?
> If they are not equal this is changing the behavior...

This gets moved to e->base.file_priv->event_list in
drm_emit_vblank_event ().  So it's doing a bit of useless work setting a
couple of pointers, but this doesn't change the behaviour.

> 
> >  	if ((seq - vblwait->request.sequence) <= (1 << 23)) {
> > -		e->event.sequence = seq;
> > -		e->event.tv_sec = now.tv_sec;
> > -		e->event.tv_usec = now.tv_usec;
> > -		drm_vblank_put(dev, pipe);
> > -		list_add_tail(&e->base.link, &e->base.file_priv->event_list);
> > -		wake_up_interruptible(&e->base.file_priv->event_wait);
> > -		vblwait->reply.sequence = seq;
> 
> Is it OK to drop this substitution?

Ahem.  No, it's not.  That's an overzealous delete key going off.  Hm.
Either the clients I was testing don't care about the reply, or the
clients I was testing always set _DRM_BLANK_NEXTONMISS.  I think it's
the latter.

Updated patch coming.

> 
> > -		trace_drm_vblank_event_delivered(current->pid, pipe,
> > -						 vblwait->request.sequence);
> > +		drm_emit_vblank_event(dev, e, seq, &now);
> >  	} else {
> > -		list_add_tail(&e->base.link, &dev->vblank_event_list);
> >  		vblwait->reply.sequence = vblwait->request.sequence;
> >  	}
> >  
> > @@ -1249,14 +1248,7 @@ void drm_handle_vblank_events(struct drm_device *dev, int crtc)
> >  		DRM_DEBUG("vblank event on %d, current %d\n",
> >  			  e->event.sequence, seq);
> >  
> > -		e->event.sequence = seq;
> > -		e->event.tv_sec = now.tv_sec;
> > -		e->event.tv_usec = now.tv_usec;
> > -		drm_vblank_put(dev, e->pipe);
> > -		list_move_tail(&e->base.link, &e->base.file_priv->event_list);
> > -		wake_up_interruptible(&e->base.file_priv->event_wait);
> > -		trace_drm_vblank_event_delivered(e->base.pid, e->pipe,
> > -						 e->event.sequence);
> > +		drm_emit_vblank_event(dev, e, seq, &now);
> >  	}
> >  
> >  	spin_unlock_irqrestore(&dev->event_lock, flags);
> > -- 



[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 490 bytes --]

[-- Attachment #2: Type: text/plain, Size: 159 bytes --]

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel

  reply	other threads:[~2011-05-01 23:55 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-04-27  6:10 [PATCH 1/3] drm: Send pending vblank events before disabling vblank christopher.halse.rogers
2011-04-27  6:10 ` [PATCH 2/3] drm: Warn if vblank state has become inconsistent christopher.halse.rogers
2011-04-27  8:38   ` Michel Dänzer
2011-04-27  6:10 ` [PATCH 3/3] drm: Factor-out drm_emit_vblank_event code christopher.halse.rogers
2011-04-27  8:36   ` Michel Dänzer
2011-04-27  8:48     ` Christopher James Halse Rogers
2011-04-27  9:06       ` Michel Dänzer
2011-04-28 20:36   ` Jesse Barnes
2011-04-29  3:57     ` [PATCH] drm: Factor-out drm_emit_vblank_event code. (v2) christopher.halse.rogers
2011-04-29  6:43       ` Michel Dänzer
2011-04-29 15:55       ` Marcin Slusarz
2011-05-01 23:55         ` Christopher James Halse Rogers [this message]
2011-05-02  0:09         ` [PATCH] drm: Factor-out drm_emit_vblank_event code. (v3) christopher.halse.rogers
2011-04-27  8:32 ` [PATCH 1/3] drm: Send pending vblank events before disabling vblank Michel Dänzer
2011-04-27  8:58   ` Christopher James Halse Rogers
2011-04-27  9:08     ` Michel Dänzer
2011-04-27  9:17       ` Christopher James Halse Rogers
2011-04-28  8:09       ` Christopher James Halse Rogers
2011-04-28 20:46         ` Jesse Barnes
2011-04-28 20:53           ` Jesse Barnes
2011-04-28 20:42   ` Jesse Barnes
2011-04-28 20:34 ` Jesse Barnes

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1304294128.25088.7.camel@Ed \
    --to=christopher.halse.rogers@canonical.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=marcin.slusarz@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.