All of lore.kernel.org
 help / color / mirror / Atom feed
From: Hans Verkuil <hverkuil@xs4all.nl>
To: Ian Molton <ian.molton@codethink.co.uk>, linux-media@vger.kernel.org
Cc: linux-kernel@lists.codethink.co.uk, g.liakhovetski@gmx.de,
	m.chehab@samsung.com, vladimir.barinov@cogentembedded.com,
	magnus.damm@gmail.com, horms@verge.net.au,
	linux-sh@vger.kernel.org
Subject: Re: [PATCH 2/4] media: rcar_vin: Ensure all in-flight buffers are returned to error state before sto
Date: Mon, 04 Aug 2014 21:22:32 +0000	[thread overview]
Message-ID: <53DFF998.1020307@xs4all.nl> (raw)
In-Reply-To: <1404812474-7627-3-git-send-email-ian.molton@codethink.co.uk>

On 07/08/2014 11:41 AM, Ian Molton wrote:
> Videobuf2 complains about buffers that are still marked ACTIVE (in use by the driver) following a call to stop_streaming().
> 
> This patch returns all active buffers to state ERROR prior to stopping.
> 
> Note: this introduces a (non fatal) race condition as the stream is not guaranteed to be stopped at this point.
> 
> Signed-off-by: Ian Molton <ian.molton@codethink.co.uk>
> Signed-off-by: William Towle <william.towle@codethink.co.uk>
> ---
>  drivers/media/platform/soc_camera/rcar_vin.c | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/drivers/media/platform/soc_camera/rcar_vin.c b/drivers/media/platform/soc_camera/rcar_vin.c
> index 7154500..06ce705 100644
> --- a/drivers/media/platform/soc_camera/rcar_vin.c
> +++ b/drivers/media/platform/soc_camera/rcar_vin.c
> @@ -513,8 +513,14 @@ static void rcar_vin_stop_streaming(struct vb2_queue *vq)
>  	struct soc_camera_host *ici = to_soc_camera_host(icd->parent);
>  	struct rcar_vin_priv *priv = ici->priv;
>  	struct list_head *buf_head, *tmp;
> +	int i;
>  
>  	spin_lock_irq(&priv->lock);
> +
> +	for (i = 0; i < vq->num_buffers; ++i)
> +		if (vq->bufs[i]->state = VB2_BUF_STATE_ACTIVE)
> +			vb2_buffer_done(vq->bufs[i], VB2_BUF_STATE_ERROR);
> +
>  	list_for_each_safe(buf_head, tmp, &priv->capture)
>  		list_del_init(buf_head);

I'm assuming all buffers that are queued to the driver via buf_queue() are
linked into priv->capture. So you would typically call vb2_buffer_done
when you are walking that list:

  	list_for_each_safe(buf_head, tmp, &priv->capture) {
		// usually you go from buf_head to the real buffer struct
		// containing a vb2_buffer struct
		vb2_buffer_done(&buf->vb, VB2_BUF_STATE_ERROR);
  		list_del_init(buf_head);
	}

Please use this rather than looking into internal vb2_queue datastructures.

Regards,

	Hans

>  	spin_unlock_irq(&priv->lock);
> 


WARNING: multiple messages have this Message-ID (diff)
From: Hans Verkuil <hverkuil@xs4all.nl>
To: Ian Molton <ian.molton@codethink.co.uk>, linux-media@vger.kernel.org
Cc: linux-kernel@lists.codethink.co.uk, g.liakhovetski@gmx.de,
	m.chehab@samsung.com, vladimir.barinov@cogentembedded.com,
	magnus.damm@gmail.com, horms@verge.net.au,
	linux-sh@vger.kernel.org
Subject: Re: [PATCH 2/4] media: rcar_vin: Ensure all in-flight buffers are returned to error state before stopping.
Date: Mon, 04 Aug 2014 23:22:32 +0200	[thread overview]
Message-ID: <53DFF998.1020307@xs4all.nl> (raw)
In-Reply-To: <1404812474-7627-3-git-send-email-ian.molton@codethink.co.uk>

On 07/08/2014 11:41 AM, Ian Molton wrote:
> Videobuf2 complains about buffers that are still marked ACTIVE (in use by the driver) following a call to stop_streaming().
> 
> This patch returns all active buffers to state ERROR prior to stopping.
> 
> Note: this introduces a (non fatal) race condition as the stream is not guaranteed to be stopped at this point.
> 
> Signed-off-by: Ian Molton <ian.molton@codethink.co.uk>
> Signed-off-by: William Towle <william.towle@codethink.co.uk>
> ---
>  drivers/media/platform/soc_camera/rcar_vin.c | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/drivers/media/platform/soc_camera/rcar_vin.c b/drivers/media/platform/soc_camera/rcar_vin.c
> index 7154500..06ce705 100644
> --- a/drivers/media/platform/soc_camera/rcar_vin.c
> +++ b/drivers/media/platform/soc_camera/rcar_vin.c
> @@ -513,8 +513,14 @@ static void rcar_vin_stop_streaming(struct vb2_queue *vq)
>  	struct soc_camera_host *ici = to_soc_camera_host(icd->parent);
>  	struct rcar_vin_priv *priv = ici->priv;
>  	struct list_head *buf_head, *tmp;
> +	int i;
>  
>  	spin_lock_irq(&priv->lock);
> +
> +	for (i = 0; i < vq->num_buffers; ++i)
> +		if (vq->bufs[i]->state == VB2_BUF_STATE_ACTIVE)
> +			vb2_buffer_done(vq->bufs[i], VB2_BUF_STATE_ERROR);
> +
>  	list_for_each_safe(buf_head, tmp, &priv->capture)
>  		list_del_init(buf_head);

I'm assuming all buffers that are queued to the driver via buf_queue() are
linked into priv->capture. So you would typically call vb2_buffer_done
when you are walking that list:

  	list_for_each_safe(buf_head, tmp, &priv->capture) {
		// usually you go from buf_head to the real buffer struct
		// containing a vb2_buffer struct
		vb2_buffer_done(&buf->vb, VB2_BUF_STATE_ERROR);
  		list_del_init(buf_head);
	}

Please use this rather than looking into internal vb2_queue datastructures.

Regards,

	Hans

>  	spin_unlock_irq(&priv->lock);
> 


  parent reply	other threads:[~2014-08-04 21:22 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-07-08  9:41 Resend: [PATCH 0/4] rcar_vin: fix soc_camera WARN_ON() issues Ian Molton
2014-07-08  9:41 ` Ian Molton
2014-07-08  9:41 ` [PATCH 1/4] media: rcar_vin: Dont aggressively retire buffers Ian Molton
2014-07-08  9:41   ` Ian Molton
2014-07-08  9:41 ` [PATCH 2/4] media: rcar_vin: Ensure all in-flight buffers are returned to error state before stoppin Ian Molton
2014-07-08  9:41   ` [PATCH 2/4] media: rcar_vin: Ensure all in-flight buffers are returned to error state before stopping Ian Molton
2014-08-04 20:39   ` [PATCH 2/4] media: rcar_vin: Ensure all in-flight buffers are returned to error state before sto Sergei Shtylyov
2014-08-04 20:39     ` [PATCH 2/4] media: rcar_vin: Ensure all in-flight buffers are returned to error state before stopping Sergei Shtylyov
2014-08-13 17:30     ` [PATCH 2/4] media: rcar_vin: Ensure all in-flight buffers are returned to error state before sto Ian Molton
2014-08-13 17:30       ` [PATCH 2/4] media: rcar_vin: Ensure all in-flight buffers are returned to error state before stopping Ian Molton
2014-08-13 17:35       ` [PATCH 2/4] media: rcar_vin: Ensure all in-flight buffers are returned to error state before sto Sergei Shtylyov
2014-08-13 17:35         ` [PATCH 2/4] media: rcar_vin: Ensure all in-flight buffers are returned to error state before stopping Sergei Shtylyov
2014-09-20  6:59       ` [PATCH 2/4] media: rcar_vin: Ensure all in-flight buffers are returned to error state before sto Guennadi Liakhovetski
2014-09-20  6:59         ` [PATCH 2/4] media: rcar_vin: Ensure all in-flight buffers are returned to error state before stopping Guennadi Liakhovetski
2014-08-04 21:22   ` Hans Verkuil [this message]
2014-08-04 21:22     ` Hans Verkuil
2015-01-18 20:32     ` [PATCH 2/4] media: rcar_vin: Ensure all in-flight buffers are returned to error state before sto Guennadi Liakhovetski
2015-01-18 20:32       ` [PATCH 2/4] media: rcar_vin: Ensure all in-flight buffers are returned to error state before stopping Guennadi Liakhovetski
2014-07-08  9:41 ` [PATCH 3/4] media: rcar_vin: Fix race condition terminating stream Ian Molton
2014-07-08  9:41   ` Ian Molton
2014-07-08 16:09   ` Sergei Shtylyov
2014-07-08 16:09     ` Sergei Shtylyov
2014-07-09  8:16     ` Simon Horman
2014-07-09  8:16       ` Simon Horman
2014-07-10 10:15     ` Ian Molton
2014-07-10 10:15       ` Ian Molton
2014-08-01 22:16       ` Sergei Shtylyov
2014-08-01 22:16         ` Sergei Shtylyov
2014-08-04 21:27   ` Hans Verkuil
2014-08-04 21:27     ` Hans Verkuil
2014-08-11 11:23     ` Ian Molton
2014-08-11 11:23       ` Ian Molton
2014-08-11 12:14       ` Hans Verkuil
2014-08-11 12:14         ` Hans Verkuil
2014-07-08  9:41 ` [PATCH 4/4] media: rcar_vin: Clean up rcar_vin_irq Ian Molton
2014-07-08  9:41   ` Ian Molton
2014-08-04 20:00 ` Resend: [PATCH 0/4] rcar_vin: fix soc_camera WARN_ON() issues Sergei Shtylyov
2014-08-04 20:00   ` Sergei Shtylyov

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=53DFF998.1020307@xs4all.nl \
    --to=hverkuil@xs4all.nl \
    --cc=g.liakhovetski@gmx.de \
    --cc=horms@verge.net.au \
    --cc=ian.molton@codethink.co.uk \
    --cc=linux-kernel@lists.codethink.co.uk \
    --cc=linux-media@vger.kernel.org \
    --cc=linux-sh@vger.kernel.org \
    --cc=m.chehab@samsung.com \
    --cc=magnus.damm@gmail.com \
    --cc=vladimir.barinov@cogentembedded.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.