linux-kernel - Re: [PATCH] ASoC: renesas: rz-ssi: Clean up on error in probe()

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <8ff70298-8a9c-4228-b064-f46ec81f6d15@tuxon.dev>
Date: Fri, 17 Jan 2025 13:51:55 +0200
From: Claudiu Beznea <claudiu.beznea@...on.dev>
To: Dan Carpenter <dan.carpenter@...aro.org>,
 Biju Das <biju.das.jz@...renesas.com>
Cc: Lad Prabhakar <prabhakar.mahadev-lad.rj@...renesas.com>,
 Liam Girdwood <lgirdwood@...il.com>, Mark Brown <broonie@...nel.org>,
 Jaroslav Kysela <perex@...ex.cz>, Takashi Iwai <tiwai@...e.com>,
 linux-sound@...r.kernel.org, linux-renesas-soc@...r.kernel.org,
 linux-kernel@...r.kernel.org, kernel-janitors@...r.kernel.org
Subject: Re: [PATCH] ASoC: renesas: rz-ssi: Clean up on error in probe()

Hi, Dan,

On 15.01.2025 08:55, Dan Carpenter wrote:
> Call rz_ssi_release_dma_channels() on these error paths to clean up from
> rz_ssi_dma_request().
> 
> Fixes: 26ac471c5354 ("ASoC: sh: rz-ssi: Add SSI DMAC support")
> Signed-off-by: Dan Carpenter <dan.carpenter@...aro.org>
> ---
>  sound/soc/renesas/rz-ssi.c | 42 ++++++++++++++++++++++++--------------
>  1 file changed, 27 insertions(+), 15 deletions(-)
> 
> diff --git a/sound/soc/renesas/rz-ssi.c b/sound/soc/renesas/rz-ssi.c
> index 3a0af4ca7ab6..4b3016282717 100644
> --- a/sound/soc/renesas/rz-ssi.c
> +++ b/sound/soc/renesas/rz-ssi.c
> @@ -1150,35 +1150,47 @@ static int rz_ssi_probe(struct platform_device *pdev)
>  		ssi->irq_rx = platform_get_irq_byname(pdev, "dma_rx");
>  		if (ssi->irq_tx == -ENXIO && ssi->irq_rx == -ENXIO) {
>  			ssi->irq_rt = platform_get_irq_byname(pdev, "dma_rt");
> -			if (ssi->irq_rt < 0)
> -				return ssi->irq_rt;
> +			if (ssi->irq_rt < 0) {
> +				ret = ssi->irq_rt;
> +				goto err_release_dma_chs;
> +			}
>  
>  			ret = devm_request_irq(dev, ssi->irq_rt,
>  					       &rz_ssi_interrupt, 0,
>  					       dev_name(dev), ssi);
> -			if (ret < 0)
> -				return dev_err_probe(dev, ret,
> -						     "irq request error (dma_rt)\n");
> +			if (ret < 0) {
> +				dev_err_probe(dev, ret,
> +					      "irq request error (dma_rt)\n");
> +				goto err_release_dma_chs;
> +			}
>  		} else {
> -			if (ssi->irq_tx < 0)
> -				return ssi->irq_tx;
> +			if (ssi->irq_tx < 0) {
> +				ret = ssi->irq_tx;
> +				goto err_release_dma_chs;
> +			}
>  
> -			if (ssi->irq_rx < 0)
> -				return ssi->irq_rx;
> +			if (ssi->irq_rx < 0) {
> +				ret = ssi->irq_rx;
> +				goto err_release_dma_chs;
> +			}
>  
>  			ret = devm_request_irq(dev, ssi->irq_tx,
>  					       &rz_ssi_interrupt, 0,
>  					       dev_name(dev), ssi);
> -			if (ret < 0)
> -				return dev_err_probe(dev, ret,
> -						"irq request error (dma_tx)\n");
> +			if (ret < 0) {
> +				dev_err_probe(dev, ret,
> +					      "irq request error (dma_tx)\n");
> +				goto err_release_dma_chs;
> +			}
>  
>  			ret = devm_request_irq(dev, ssi->irq_rx,
>  					       &rz_ssi_interrupt, 0,
>  					       dev_name(dev), ssi);
> -			if (ret < 0)
> -				return dev_err_probe(dev, ret,
> -						"irq request error (dma_rx)\n");
> +			if (ret < 0) {
> +				dev_err_probe(dev, ret,
> +					      "irq request error (dma_rx)\n");
> +				goto err_release_dma_chs;
> +			}
>  		}
>  	}

The code block ending here is entered only if !rz_ssi_is_dma_enabled(). If
that is true there are no DMA channel to be released. Maybe better would be
to move this code block on the failure path of the rz_ssi_dma_request() as
also proposed here:
https://lore.kernel.org/all/CAMuHMdU+_NuLp2FuwwcLfJRe2ssMtp=z7fqcsANgYfFehTNJGg@mail.gmail.com/

Thank you,
Claudiu

>