linux-media.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Philipp Zabel <p.zabel@pengutronix.de>
To: Fabio Estevam <fabio.estevam@freescale.com>
Cc: k.debski@samsung.com, m.chehab@samsung.com,
	kernel@pengutronix.de, linux-media@vger.kernel.org
Subject: Re: [PATCH v4 1/3] [media] coda: Fix error paths
Date: Wed, 24 Jul 2013 11:47:02 +0200	[thread overview]
Message-ID: <1374659222.3929.4.camel@pizza.hi.pengutronix.de> (raw)
In-Reply-To: <1374602690-12842-1-git-send-email-fabio.estevam@freescale.com>

Hi Fabio,

Am Dienstag, den 23.07.2013, 15:04 -0300 schrieb Fabio Estevam:
> Some resources were not being released in the error path and some were released
> in the incorrect order.
> 
> Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
> ---
> Changes since v3:
> - Only disable the clocks after v4l2_m2m_ctx_release()
> Changes since v2:
> - Newly introduced in this series
> 
>  drivers/media/platform/coda.c | 38 ++++++++++++++++++++++++--------------
>  1 file changed, 24 insertions(+), 14 deletions(-)
> 
> diff --git a/drivers/media/platform/coda.c b/drivers/media/platform/coda.c
> index df4ada88..9384f02 100644
> --- a/drivers/media/platform/coda.c
> +++ b/drivers/media/platform/coda.c
> @@ -1514,21 +1514,26 @@ static int coda_open(struct file *file)
>  	int ret = 0;
>  	int idx;
>  
> -	idx = coda_next_free_instance(dev);
> -	if (idx >= CODA_MAX_INSTANCES)
> -		return -EBUSY;
> -	set_bit(idx, &dev->instance_mask);
> -
>  	ctx = kzalloc(sizeof *ctx, GFP_KERNEL);
>  	if (!ctx)
>  		return -ENOMEM;
>  
> +	idx = coda_next_free_instance(dev);
> +	if (idx >= CODA_MAX_INSTANCES) {
> +		ret =  -EBUSY;
> +		goto err_coda_max;
> +	}
> +	set_bit(idx, &dev->instance_mask);
> +
>  	v4l2_fh_init(&ctx->fh, video_devdata(file));
>  	file->private_data = &ctx->fh;
>  	v4l2_fh_add(&ctx->fh);
>  	ctx->dev = dev;
>  	ctx->idx = idx;
>  
> +	clk_prepare_enable(dev->clk_per);
> +	clk_prepare_enable(dev->clk_ahb);
> +
>  	set_default_params(ctx);
>  	ctx->m2m_ctx = v4l2_m2m_ctx_init(dev->m2m_dev, ctx,
>  					 &coda_queue_init);
> @@ -1537,12 +1542,12 @@ static int coda_open(struct file *file)
>  
>  		v4l2_err(&dev->v4l2_dev, "%s return error (%d)\n",
>  			 __func__, ret);
> -		goto err;
> +		goto err_ctx_init;
>  	}
>  	ret = coda_ctrls_setup(ctx);
>  	if (ret) {
>  		v4l2_err(&dev->v4l2_dev, "failed to setup coda controls\n");
> -		goto err;
> +		goto err_ctrls_setup;
>  	}
>  
>  	ctx->fh.ctrl_handler = &ctx->ctrls;
> @@ -1552,24 +1557,29 @@ static int coda_open(struct file *file)
>  	if (!ctx->parabuf.vaddr) {
>  		v4l2_err(&dev->v4l2_dev, "failed to allocate parabuf");
>  		ret = -ENOMEM;
> -		goto err;
> +		goto err_dma_alloc;
>  	}
>  
>  	coda_lock(ctx);
>  	list_add(&ctx->list, &dev->instances);
>  	coda_unlock(ctx);
>  
> -	clk_prepare_enable(dev->clk_per);
> -	clk_prepare_enable(dev->clk_ahb);
> -
>  	v4l2_dbg(1, coda_debug, &dev->v4l2_dev, "Created instance %d (%p)\n",
>  		 ctx->idx, ctx);
>  
>  	return 0;
>  
> -err:
> +err_dma_alloc:
> +	v4l2_ctrl_handler_free(&ctx->ctrls);
> +err_ctrls_setup:
> +	v4l2_m2m_ctx_release(ctx->m2m_ctx);
> +err_ctx_init:
> +	clk_disable_unprepare(dev->clk_ahb);
> +	clk_disable_unprepare(dev->clk_per);
>  	v4l2_fh_del(&ctx->fh);
>  	v4l2_fh_exit(&ctx->fh);
> +	clear_bit(ctx->idx, &dev->instance_mask);
> +err_coda_max:
>  	kfree(ctx);
>  	return ret;
>  }
> @@ -1588,10 +1598,10 @@ static int coda_release(struct file *file)
>  
>  	dma_free_coherent(&dev->plat_dev->dev, CODA_PARA_BUF_SIZE,
>  		ctx->parabuf.vaddr, ctx->parabuf.paddr);
> -	v4l2_m2m_ctx_release(ctx->m2m_ctx);
>  	v4l2_ctrl_handler_free(&ctx->ctrls);
> -	clk_disable_unprepare(dev->clk_per);
> +	v4l2_m2m_ctx_release(ctx->m2m_ctx);
>  	clk_disable_unprepare(dev->clk_ahb);
> +	clk_disable_unprepare(dev->clk_per);
>  	v4l2_fh_del(&ctx->fh);
>  	v4l2_fh_exit(&ctx->fh);
>  	clear_bit(ctx->idx, &dev->instance_mask);

this looks better, thanks. Could you rebase the patches onto Kamil's
http://git.linuxtv.org/kdebski/media.git/shortlog/refs/heads/new-for-3.12
branch?

regards
Philipp


  parent reply	other threads:[~2013-07-24  9:48 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-07-23 18:04 [PATCH v4 1/3] [media] coda: Fix error paths Fabio Estevam
2013-07-23 18:04 ` [PATCH v4 2/3] [media] coda: Check the return value from clk_prepare_enable() Fabio Estevam
2013-07-23 18:04 ` [PATCH v4 3/3] [media] coda: No need to check the return value of platform_get_resource() Fabio Estevam
2013-07-24  9:47 ` Philipp Zabel [this message]
2013-08-01 15:27   ` [PATCH v4 1/3] [media] coda: Fix error paths Kamil Debski

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=1374659222.3929.4.camel@pizza.hi.pengutronix.de \
    --to=p.zabel@pengutronix.de \
    --cc=fabio.estevam@freescale.com \
    --cc=k.debski@samsung.com \
    --cc=kernel@pengutronix.de \
    --cc=linux-media@vger.kernel.org \
    --cc=m.chehab@samsung.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).