Alsa-Devel Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Takashi Iwai <tiwai@suse.de>
To: "Subhransu S. Prusty" <subhransu.s.prusty@intel.com>
Cc: patches.audio@intel.com, Vinod Koul <vinod.koul@intel.com>,
	alsa-devel@alsa-project.org, broonie@kernel.org,
	lgirdwood@gmail.com
Subject: Re: [PATCH 03/15] ASoC: hdac_hdmi - Use list to add pins and converters
Date: Tue, 01 Dec 2015 19:52:14 +0100	[thread overview]
Message-ID: <s5hbnaa57dt.wl-tiwai@suse.de> (raw)
In-Reply-To: <20151201225050.GE5624@subhransu-desktop>

On Tue, 01 Dec 2015 23:50:50 +0100,
Subhransu S. Prusty wrote:
> 
> On Tue, Dec 01, 2015 at 01:47:04PM +0100, Takashi Iwai wrote:
> > On Tue, 01 Dec 2015 18:46:59 +0100,
> > Subhransu S. Prusty wrote:
> > > 
> > > +static int hdac_hdmi_add_cvt(struct hdac_ext_device *edev, hda_nid_t nid)
> > > +{
> > > +	struct hdac_hdmi_priv *hdmi = edev->private_data;
> > > +	struct hdac_hdmi_cvt *cvt;
> > > +
> > > +	cvt = devm_kzalloc(&edev->hdac.dev, sizeof(*cvt), GFP_KERNEL);
> > 
> > Be careful about devm_*() usage here.  The devres resources may be
> > freed before the release callback is called for the assigned device
> > object.  See drivers/base/core.c:device_release().
> > 
> > For the top-level object, usually it's OK, because the top-level
> > object itself won't be released but only the driver is detached.
> > Then devres_release_all() is called in device_release_driver() while
> > the device itself is still present.
> 
> Not sure if I understood it correctly. But as the devres_release_all is
> called after the driver remove and the above resource is not expected to
> be used outside the scope of driver.

devres_release_all() is called *before* calling the device's release
callback, which is usually triggered by put_device().  So it depends
on how you call the device removal whether it's safe or not.

> So should't deleting the list entries
> be good enough during driver remove?

It depends on how the allocated data is accessed.  I guess it won't be
a problem in this case, but in general, you have to be careful about
devm_*() usage.  It can't be used always blindly.

> > I'm not sure whether this would be actually OK with hdac_device.  You
> > need to double-check, not only check whether the driver appears to
> > work, but track all object lifetime properly.
> 
> Will testing driver load/unload be helpful here with CONFIG_DEBUG_DEVRES
> enabled?

Maybe.  But at best you should read the code and follow the flow
closely by yourself.


Takashi

> 
> Regards,
> Subhransu
> > 
> > 
> > Takashi
> > 
> > > +	if (!cvt)
> > > +		return -ENOMEM;
> > > +
> > > +	cvt->nid = nid;
> > > +
> > > +	list_add_tail(&cvt->head, &hdmi->cvt_list);
> > > +	hdmi->num_cvt++;
> > > +
> > > +	return hdac_hdmi_query_cvt_params(&edev->hdac, cvt);
> > > +}
> > > +
> > > +static int hdac_hdmi_add_pin(struct hdac_ext_device *edev, hda_nid_t nid)
> > > +{
> > > +	struct hdac_hdmi_priv *hdmi = edev->private_data;
> > > +	struct hdac_hdmi_pin *pin;
> > > +
> > > +	pin = devm_kzalloc(&edev->hdac.dev, sizeof(*pin), GFP_KERNEL);
> > > +	if (!pin)
> > > +		return -ENOMEM;
> > > +
> > > +	pin->nid = nid;
> > > +
> > > +	list_add_tail(&pin->head, &hdmi->pin_list);
> > > +	hdmi->num_pin++;
> > > +
> > > +	return 0;
> > >  }
> > >  
> > >  /*
> > > @@ -414,7 +459,7 @@ static int hdac_hdmi_parse_and_map_nid(struct hdac_ext_device *edev)
> > >  	int i;
> > >  	struct hdac_device *hdac = &edev->hdac;
> > >  	struct hdac_hdmi_priv *hdmi = edev->private_data;
> > > -	int cvt_nid = 0, pin_nid = 0;
> > > +	int ret;
> > >  
> > >  	hdac->num_nodes = snd_hdac_get_sub_nodes(hdac, hdac->afg, &nid);
> > >  	if (!nid || hdac->num_nodes <= 0) {
> > > @@ -437,29 +482,25 @@ static int hdac_hdmi_parse_and_map_nid(struct hdac_ext_device *edev)
> > >  		switch (type) {
> > >  
> > >  		case AC_WID_AUD_OUT:
> > > -			hdmi->cvt_nid[cvt_nid] = nid;
> > > -			cvt_nid++;
> > > +			ret = hdac_hdmi_add_cvt(edev, nid);
> > > +			if (ret < 0)
> > > +				return ret;
> > >  			break;
> > >  
> > >  		case AC_WID_PIN:
> > > -			hdmi->pin_nid[pin_nid] = nid;
> > > -			pin_nid++;
> > > +			ret = hdac_hdmi_add_pin(edev, nid);
> > > +			if (ret < 0)
> > > +				return ret;
> > >  			break;
> > >  		}
> > >  	}
> > >  
> > >  	hdac->end_nid = nid;
> > >  
> > > -	if (!pin_nid || !cvt_nid)
> > > +	if (!hdmi->num_pin || !hdmi->num_cvt)
> > >  		return -EIO;
> > >  
> > > -	/*
> > > -	 * Currently on board only 1 pin and 1 converter is enabled for
> > > -	 * simplification, more will be added eventually
> > > -	 * So using fixed map for dai_id:pin:cvt
> > > -	 */
> > > -	return hdac_hdmi_init_dai_map(edev, &hdmi->dai_map[0], hdmi->pin_nid[0],
> > > -			hdmi->cvt_nid[0], 0);
> > > +	return hdac_hdmi_init_dai_map(edev);
> > >  }
> > >  
> > >  static int hdmi_codec_probe(struct snd_soc_codec *codec)
> > > @@ -543,6 +584,9 @@ static int hdac_hdmi_dev_probe(struct hdac_ext_device *edev)
> > >  
> > >  	dev_set_drvdata(&codec->dev, edev);
> > >  
> > > +	INIT_LIST_HEAD(&hdmi_priv->pin_list);
> > > +	INIT_LIST_HEAD(&hdmi_priv->cvt_list);
> > > +
> > >  	ret = hdac_hdmi_parse_and_map_nid(edev);
> > >  	if (ret < 0)
> > >  		return ret;
> > > -- 
> > > 1.9.1
> > > 
> 
> -- 
> 

  reply	other threads:[~2015-12-01 18:52 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-12-01 17:42 [PATCH 00/15] ASoC: hdac_hdmi: Add DP & notification support Subhransu S. Prusty
2015-12-01 17:46 ` [PATCH 01/15] ASoC: hdac_hdmi: Fix to check num nodes correctly Subhransu S. Prusty
2015-12-01 12:27   ` Takashi Iwai
2015-12-01 19:08     ` Subhransu S. Prusty
2015-12-01 17:46   ` [PATCH 02/15] ASoC: hdac_hdmi: Fix to warn instead of err for no connected nids Subhransu S. Prusty
2016-01-08 13:44     ` Applied "ASoC: hdac_hdmi: Fix to warn instead of err for no connected nids" to the asoc tree Mark Brown
2015-12-01 17:46   ` [PATCH 03/15] ASoC: hdac_hdmi - Use list to add pins and converters Subhransu S. Prusty
2015-12-01 12:47     ` Takashi Iwai
2015-12-01 22:50       ` Subhransu S. Prusty
2015-12-01 18:52         ` Takashi Iwai [this message]
2015-12-02  9:31           ` Takashi Iwai
2015-12-03 10:36             ` Subhransu S. Prusty
2015-12-01 17:47   ` [PATCH 04/15] ALSA: hda - Add helper to read eld data Subhransu S. Prusty
2015-12-01 12:39     ` Takashi Iwai
2015-12-01 19:12       ` Subhransu S. Prusty
2015-12-01 13:55         ` Takashi Iwai
2015-12-01 17:47   ` [PATCH 05/15] ASoC: hdac_hdmi: Add hotplug notification and read eld Subhransu S. Prusty
2015-12-01 12:37     ` Takashi Iwai
2015-12-01 20:14       ` Subhransu S. Prusty
2015-12-01 17:47   ` [PATCH 06/15] ALSA: pcm: Add DRM helper to set constraint for format Subhransu S. Prusty
2015-12-01 12:55     ` Takashi Iwai
2015-12-02 12:27       ` Subhransu S. Prusty
2015-12-02  9:23         ` Takashi Iwai
2015-12-01 17:47   ` [PATCH 07/15] ASoC: hdac_hdmi: Apply constraints based on ELD Subhransu S. Prusty
2015-12-01 17:47   ` [PATCH 08/15] ASoC: hdac_hdmi: Enable DP1.2 and all converters/pins Subhransu S. Prusty
2015-12-01 17:47   ` [PATCH 09/15] ASoC: hdac_hdmi - create dais based on number of streams Subhransu S. Prusty
2015-12-01 17:47   ` [PATCH 10/15] ASoC: hdac_hdmi: Create widget/route based on nodes enumerated Subhransu S. Prusty
2015-12-01 17:47   ` [PATCH 11/15] ASoC: hdac_hdmi: Assign pin for stream based on dapm connection Subhransu S. Prusty
2015-12-01 17:47   ` [PATCH 12/15] drm/edid: Add API to help find connection type Subhransu S. Prusty
2015-12-02  9:53     ` Jani Nikula
2015-12-02 17:07       ` Thierry Reding
2015-12-03 16:08         ` [alsa-devel] " Subhransu S. Prusty
2015-12-03 11:09           ` Jani Nikula
2015-12-03 11:21             ` Thierry Reding
2015-12-03 17:14               ` Subhransu S. Prusty
2015-12-02 17:16       ` Subhransu S. Prusty
2015-12-01 17:47   ` [PATCH 13/15] ASoC: hdac_hdmi: Add infoframe support for dp audio Subhransu S. Prusty
2015-12-01 17:47   ` [PATCH 14/15] ASoC: hdac_hdmi: Add codec suspend/resume handler Subhransu S. Prusty
2015-12-01 17:47   ` [PATCH 15/15] ASoC: hdac_hdmi: Keep display active while enumerating codec Subhransu S. Prusty
2015-12-01 12:57     ` Takashi Iwai
2015-12-01 21:48       ` Subhransu S. Prusty
2015-12-01 16:33         ` Takashi Iwai
2015-12-01 16:48           ` Babu, Ramesh
2015-12-01 17:08             ` Takashi Iwai
2015-12-02 15:37               ` Subhransu S. Prusty
2015-12-02 10:24                 ` Takashi Iwai
2015-12-02 10:31                   ` Takashi Iwai
2015-12-02 16:23                   ` Subhransu S. Prusty

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=s5hbnaa57dt.wl-tiwai@suse.de \
    --to=tiwai@suse.de \
    --cc=alsa-devel@alsa-project.org \
    --cc=broonie@kernel.org \
    --cc=lgirdwood@gmail.com \
    --cc=patches.audio@intel.com \
    --cc=subhransu.s.prusty@intel.com \
    --cc=vinod.koul@intel.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