[alsa-devel] [PATCH] ASoC: tegra: Convert to managed resources

[Alexandre Courbot]

On Mon, Aug 17, 2015 at 6:32 PM, Vaishali Thakkar
<vthakkar1994 at gmail.com> wrote:
> On Mon, Aug 17, 2015 at 12:47 PM, Alexandre Courbot <gnurou at gmail.com> wrote:
>> On Thu, Aug 13, 2015 at 4:29 PM, Vaishali Thakkar
>> <vthakkar1994 at gmail.com> wrote:
>>> Use managed resource functions devm_clk_put and
>>> devm_snd_soc_register_component to simplify error handling.
>>>
>>> To be compatible with the change various gotos are replaced
>>> with direct returns, and unneeded labels are dropped.
>>>
>>> Signed-off-by: Vaishali Thakkar <vthakkar1994 at gmail.com>
>>> ---
>>>  sound/soc/tegra/tegra20_spdif.c | 37 +++++++++++++------------------------
>>>  1 file changed, 13 insertions(+), 24 deletions(-)
>>>
>>> diff --git a/sound/soc/tegra/tegra20_spdif.c b/sound/soc/tegra/tegra20_spdif.c
>>> index 9141477..f69b2e4 100644
>>> --- a/sound/soc/tegra/tegra20_spdif.c
>>> +++ b/sound/soc/tegra/tegra20_spdif.c
>>> @@ -273,45 +273,40 @@ static int tegra20_spdif_platform_probe(struct platform_device *pdev)
>>>                              GFP_KERNEL);
>>>         if (!spdif) {
>>>                 dev_err(&pdev->dev, "Can't allocate tegra20_spdif\n");
>>> -               ret = -ENOMEM;
>>> -               goto err;
>>> +               return -ENOMEM;
>>>         }
>>>         dev_set_drvdata(&pdev->dev, spdif);
>>>
>>> -       spdif->clk_spdif_out = clk_get(&pdev->dev, "spdif_out");
>>> +       spdif->clk_spdif_out = devm_clk_get(&pdev->dev, "spdif_out");
>>>         if (IS_ERR(spdif->clk_spdif_out)) {
>>>                 pr_err("Can't retrieve spdif clock\n");
>>>                 ret = PTR_ERR(spdif->clk_spdif_out);
>>> -               goto err;
>>> +               return ret;
>>
>> Maybe do "return PTR_ERR(spdif->clk_spdif_out);" for consistency with
>> the other error cases of this function?
>>
>>>         }
>>>
>>>         mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
>>>         if (!mem) {
>>>                 dev_err(&pdev->dev, "No memory resource\n");
>>> -               ret = -ENODEV;
>>> -               goto err_clk_put;
>>> +               return -ENODEV;
>>>         }
>>>
>>>         dmareq = platform_get_resource(pdev, IORESOURCE_DMA, 0);
>>>         if (!dmareq) {
>>>                 dev_err(&pdev->dev, "No DMA resource\n");
>>> -               ret = -ENODEV;
>>> -               goto err_clk_put;
>>> +               return -ENODEV;
>>>         }
>>>
>>>         memregion = devm_request_mem_region(&pdev->dev, mem->start,
>>>                                             resource_size(mem), DRV_NAME);
>>>         if (!memregion) {
>>>                 dev_err(&pdev->dev, "Memory region already claimed\n");
>>> -               ret = -EBUSY;
>>> -               goto err_clk_put;
>>> +               return -EBUSY;
>>>         }
>>>
>>>         regs = devm_ioremap(&pdev->dev, mem->start, resource_size(mem));
>>>         if (!regs) {
>>>                 dev_err(&pdev->dev, "ioremap failed\n");
>>> -               ret = -ENOMEM;
>>> -               goto err_clk_put;
>>> +               return -ENOMEM;
>>>         }
>>>
>>>         spdif->regmap = devm_regmap_init_mmio(&pdev->dev, regs,
>>> @@ -319,7 +314,7 @@ static int tegra20_spdif_platform_probe(struct platform_device *pdev)
>>>         if (IS_ERR(spdif->regmap)) {
>>>                 dev_err(&pdev->dev, "regmap init failed\n");
>>>                 ret = PTR_ERR(spdif->regmap);
>>> -               goto err_clk_put;
>>> +               return ret;
>>
>> Same here.
>
> Actually people prefer to write this way when they are calling PTR_ERR
> more than one time for the same value. But as for this file at both places
> we are calling PTR_ERR for different values, may be we can directly call
> it in return.

Ok, I don't feel too strongly about this, so your call.

>
>>>         }
>>>
>>>         spdif->playback_dma_data.addr = mem->start + TEGRA20_SPDIF_DATA_OUT;
>>> @@ -334,8 +329,9 @@ static int tegra20_spdif_platform_probe(struct platform_device *pdev)
>>>                         goto err_pm_disable;
>>>         }
>>>
>>> -       ret = snd_soc_register_component(&pdev->dev, &tegra20_spdif_component,
>>> -                                  &tegra20_spdif_dai, 1);
>>> +       ret = devm_snd_soc_register_component(&pdev->dev,
>>> +                                             &tegra20_spdif_component,
>>> +                                             &tegra20_spdif_dai, 1);
>>>         if (ret) {
>>>                 dev_err(&pdev->dev, "Could not register DAI: %d\n", ret);
>>>                 ret = -ENOMEM;
>>> @@ -345,21 +341,17 @@ static int tegra20_spdif_platform_probe(struct platform_device *pdev)
>>>         ret = tegra_pcm_platform_register(&pdev->dev);
>>>         if (ret) {
>>>                 dev_err(&pdev->dev, "Could not register PCM: %d\n", ret);
>>> -               goto err_unregister_component;
>>> +               return ret;
>>
>> In the previous code, PM cleanup was also performed after the
>> component was unregistered. If you return directly, this is not
>> performed anymore - I think you should "goto err_suspend;" here.
>> This will change the ordering of cleanup operations though - e.g.
>> snd_soc_unregister_component() will now be called *after* PM cleanup.
>> Are things still working ok after this? (I suppose they do considering
>> how simple the PM ops are, but it might be worth testing).
>
> I think you are right. I missed that. But now thing is , this patch is already
> applied here:
> https://git.kernel.org/cgit/linux/kernel/git/broonie/sound.git/commit/?h=topic/tegra

Ah, that will teach me to check my patches queue more often. :(

>
> I am not sure if now I can change version 2 with the changes you suggested
> or not. Although I will not be able to test it  after changing 'goto
> err_suspend'
> as I don't have hardware but may be someone else can test it.

I think you can send a fixup patch since Mark already merged this one,
this is an error code path (which by definition should not be taken
too often), and there should not be any resulting breakage.