[v2,1/2] ALSA: hda/intel: Fix error handling in azx_probe()

Commit Message

Maarten Lankhorst July 18, 2023, 11:57 a.m. UTC

Hey,

On 2023-07-18 11:57, Takashi Iwai wrote:
> On Tue, 18 Jul 2023 10:45:20 +0200,
> Maarten Lankhorst wrote:
>> Now that we can use -EPROBE_DEFER, it's no longer required to spin off
>> the snd_hdac_i915_init into a workqueue.
>>
>> Use the -EPROBE_DEFER mechanism instead, which must be returned in the
>> probe function.
>>
>> Signed-off-by: Maarten Lankhorst<[e-mail]>
>> ---
>>   sound/pci/hda/hda_intel.c | 58 +++++++++++++++++++++------------------
>>   1 file changed, 31 insertions(+), 27 deletions(-)
>>
>> diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c
>> index 5af1138e745bc..d40345a0088d8 100644
>> --- a/sound/pci/hda/hda_intel.c
>> +++ b/sound/pci/hda/hda_intel.c
>> @@ -213,6 +213,7 @@ MODULE_DESCRIPTION("Intel HDA driver");
>>   #endif
>>   #endif
>>   
>> +static DECLARE_BITMAP(probed_devs, SNDRV_CARDS);
>>   
>>   /*
>>    */
>> @@ -2094,8 +2095,6 @@ static const struct hda_controller_ops pci_hda_ops = {
>>   	.position_check = azx_position_check,
>>   };
>>   
>> -static DECLARE_BITMAP(probed_devs, SNDRV_CARDS);
>> -
>>   static int azx_probe(struct pci_dev *pci,
>>   		     const struct pci_device_id *pci_id)
>>   {
> Any specific reason to move the definition?  Otherwise let's
> concentrate on the needed change.

Originally I moved the chunk to azx_create(), but I felt it was the 
wrong place. Fixed now.

>
>> @@ -2174,7 +2173,36 @@ static int azx_probe(struct pci_dev *pci,
>>   	}
>>   #endif /* CONFIG_SND_HDA_PATCH_LOADER */
>>   
>> -#ifndef CONFIG_SND_HDA_I915
>> +#ifdef CONFIG_SND_HDA_I915
>> +	/* bind with i915 if needed */
>> +	if (chip->driver_caps & AZX_DCAPS_I915_COMPONENT) {
>> +		err = snd_hdac_i915_init(azx_bus(chip), false);
>> +		if (err < 0) {
>> +			/* if the controller is bound only with HDMI/DP
>> +			 * (for HSW and BDW), we need to abort the probe;
>> +			 * for other chips, still continue probing as other
>> +			 * codecs can be on the same link.
>> +			 */
>> +			if (CONTROLLER_IN_GPU(pci)) {
>> +				if (err != -EPROBE_DEFER)
>> +					dev_err(card->dev,
>> +						"HSW/BDW HD-audio HDMI/DP requires binding with gfx driver\n");
>> +
>> +				clear_bit(chip->dev_index, probed_devs);
>> +				pci_set_drvdata(pci, NULL);
>> +				snd_device_free(card, chip);
>> +				return err;
> This may leak resources, I'm afraid.
>
> Here you just need to "goto out_free;" instead of manual resource
> releases, which eventually calls snd_card_free(), and that's all.
>
> (Though, pci_set_drvdata(pci, NULL) might be still missing; but it's
>   not only for this change, and we'll need to address it in another
>   patch.)

git am --scissors patch below for cleanup of error path.

It seems azx_free will unregister from vga_switcheroo safely.

------------>8---

Make sure azx is freed after azx_create() succeeded and an error was
encountered.

Signed-off-by: Maarten Lankhorst<[e-mail]>
---
  sound/pci/hda/hda_intel.c | 7 +++++--
  1 file changed, 5 insertions(+), 2 deletions(-)

Patch

diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c
index 5af1138e745bc..196ca76ac43ad 100644
--- a/sound/pci/hda/hda_intel.c
+++ b/sound/pci/hda/hda_intel.c
@@ -2150,7 +2150,7 @@  static int azx_probe(struct pci_dev *pci,
  	err = register_vga_switcheroo(chip);
  	if (err < 0) {
  		dev_err(card->dev, "Error registering vga_switcheroo client\n");
-		goto out_free;
+		goto out_azx_free;
  	}
  
  	if (check_hdmi_disabled(pci)) {
@@ -2169,7 +2169,7 @@  static int azx_probe(struct pci_dev *pci,
  					      &pci->dev, GFP_KERNEL, card,
  					      azx_firmware_cb);
  		if (err < 0)
-			goto out_free;
+			goto out_azx_free;
  		schedule_probe = false; /* continued in azx_firmware_cb() */
  	}
  #endif /* CONFIG_SND_HDA_PATCH_LOADER */
@@ -2187,6 +2187,9 @@  static int azx_probe(struct pci_dev *pci,
  		complete_all(&hda->probe_wait);
  	return 0;
  
+out_azx_free:
+	azx_free(chip);
+	pci_set_drvdata(pci, NULL);
  out_free:
  	snd_card_free(card);
  	return err;