DMA Engine development
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Pan Chuang" <panchuang@vivo.com>
Cc: Frank.Li@kernel.org, dmaengine@vger.kernel.org, vkoul@kernel.org
Subject: Re: [PATCH 18/26] dmaengine: stm32-stm32-dma: Remove redundant dev_err()/dev_err_probe()
Date: Thu, 09 Jul 2026 14:30:02 +0000	[thread overview]
Message-ID: <20260709143002.B89541F000E9@smtp.kernel.org> (raw)
In-Reply-To: <20260709135846.97972-19-panchuang@vivo.com>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] Dangling pointer in `irqaction->name` leads to a Use-After-Free read when accessing `/proc/interrupts` on probe failure due to premature device name deallocation.
--

--- Patch [18]: [PATCH 18/26] dmaengine: stm32-stm32-dma: Remove redundant dev_err()/dev_err_probe() ---
commit deae2b976db4b82d7a92a5e0ad3b7fc0fc39d45b
Author: Pan Chuang <panchuang@vivo.com>

dmaengine: stm32-stm32-dma: Remove redundant dev_err()/dev_err_probe()

This commit removes redundant driver-specific error logging when
devm_request_irq() fails. The core now logs detailed error messages
automatically.

> diff --git a/drivers/dma/stm32/stm32-dma.c b/drivers/dma/stm32/stm32-dma.c
> index d3ad78562a140..51cb2328a8e19 100644
> --- a/drivers/dma/stm32/stm32-dma.c
> +++ b/drivers/dma/stm32/stm32-dma.c
> @@ -1669,12 +1669,8 @@ static int stm32_dma_probe(struct platform_device *pdev)
>  		ret = devm_request_irq(&pdev->dev, chan->irq,
>  				       stm32_dma_chan_irq, 0,
>  				       dev_name(chan2dev(chan)), chan);
> -		if (ret) {
> -			dev_err(&pdev->dev,
> -				"request_irq failed with err %d channel %d\n",
> -				ret, i);
> +		if (ret)
>  			goto err_unregister;

[Severity: High]
This is a pre-existing issue, but does this error path leave a dangling
pointer in the IRQ subsystem?

In stm32_dma_probe(), the devm_request_irq() call uses
dev_name(chan2dev(chan)) as the IRQ name. The IRQ subsystem stores this
pointer directly in irqaction->name without copying the string.

If the probe function fails and jumps to err_unregister,
dma_async_device_unregister() is called:

err_unregister:
	dma_async_device_unregister(dd);

This unregisters the channel devices and frees their kobject name strings.
Since the IRQs are devm-managed, they remain active until the probe function
completely returns and devres_release_all() cleans them up.

During this window, can a concurrent read of /proc/interrupts access the
freed device name string?

> -		}
>  	}
>  
>  	ret = of_dma_controller_register(pdev->dev.of_node,

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260709135846.97972-1-panchuang@vivo.com?part=18

  reply	other threads:[~2026-07-09 14:30 UTC|newest]

Thread overview: 56+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-07-09 13:58 [PATCH 00/26] dmaengine: Remove redundant error messages on IRQ request failure Pan Chuang
2026-07-09 13:58 ` [PATCH 01/26] dmaengine: fsl-edma-main: Remove redundant dev_err()/dev_err_probe() Pan Chuang
2026-07-09 14:12   ` sashiko-bot
2026-07-09 14:33   ` Frank Li
2026-07-09 13:58 ` [PATCH 02/26] dmaengine: fsl-qdma: " Pan Chuang
2026-07-09 14:08   ` sashiko-bot
2026-07-09 14:34   ` Frank Li
2026-07-09 13:58 ` [PATCH 03/26] dmaengine: loongson-loongson1-apb-dma: " Pan Chuang
2026-07-09 14:11   ` sashiko-bot
2026-07-09 13:58 ` [PATCH 04/26] dmaengine: mediatek-mtk-cqdma: " Pan Chuang
2026-07-09 14:10   ` sashiko-bot
2026-07-09 13:58 ` [PATCH 05/26] dmaengine: mediatek-mtk-hsdma: " Pan Chuang
2026-07-09 14:10   ` sashiko-bot
2026-07-09 13:58 ` [PATCH 06/26] dmaengine: mmp_pdma: " Pan Chuang
2026-07-09 14:08   ` sashiko-bot
2026-07-09 13:58 ` [PATCH 07/26] dmaengine: moxart-dma: " Pan Chuang
2026-07-09 14:16   ` sashiko-bot
2026-07-09 13:58 ` [PATCH 08/26] dmaengine: owl-dma: " Pan Chuang
2026-07-09 14:16   ` sashiko-bot
2026-07-09 13:58 ` [PATCH 09/26] dmaengine: pxa_dma: " Pan Chuang
2026-07-09 14:20   ` sashiko-bot
2026-07-09 13:58 ` [PATCH 10/26] dmaengine: qcom-gpi: " Pan Chuang
2026-07-09 14:19   ` sashiko-bot
2026-07-09 13:58 ` [PATCH 11/26] dmaengine: sf-pdma-sf-pdma: " Pan Chuang
2026-07-09 14:18   ` sashiko-bot
2026-07-09 13:58 ` [PATCH 12/26] dmaengine: sh-rcar-dmac: " Pan Chuang
2026-07-09 14:23   ` sashiko-bot
2026-07-09 14:31   ` Geert Uytterhoeven
2026-07-09 13:58 ` [PATCH 13/26] dmaengine: sh-rz-dmac: " Pan Chuang
2026-07-09 14:32   ` Geert Uytterhoeven
2026-07-09 13:58 ` [PATCH 14/26] dmaengine: sh-shdmac: " Pan Chuang
2026-07-09 14:24   ` sashiko-bot
2026-07-09 13:58 ` [PATCH 15/26] dmaengine: sh-usb-dmac: " Pan Chuang
2026-07-09 14:27   ` sashiko-bot
2026-07-09 14:33   ` Geert Uytterhoeven
2026-07-09 13:58 ` [PATCH 16/26] dmaengine: sprd-dma: " Pan Chuang
2026-07-09 14:27   ` sashiko-bot
2026-07-09 13:58 ` [PATCH 17/26] dmaengine: st_fdma: " Pan Chuang
2026-07-09 14:30   ` sashiko-bot
2026-07-09 13:58 ` [PATCH 18/26] dmaengine: stm32-stm32-dma: " Pan Chuang
2026-07-09 14:30   ` sashiko-bot [this message]
2026-07-09 13:58 ` [PATCH 19/26] dmaengine: stm32-stm32-dma3: " Pan Chuang
2026-07-09 14:33   ` sashiko-bot
2026-07-09 13:58 ` [PATCH 20/26] dmaengine: stm32-stm32-mdma: " Pan Chuang
2026-07-09 14:32   ` sashiko-bot
2026-07-09 13:58 ` [PATCH 21/26] dmaengine: sun4i-dma: " Pan Chuang
2026-07-09 14:35   ` sashiko-bot
2026-07-09 13:58 ` [PATCH 22/26] dmaengine: sun6i-dma: " Pan Chuang
2026-07-09 13:58 ` [PATCH 23/26] dmaengine: tegra20-apb-dma: " Pan Chuang
2026-07-09 13:58 ` [PATCH 24/26] dmaengine: ti-edma: " Pan Chuang
2026-07-09 14:38   ` sashiko-bot
2026-07-09 13:58 ` [PATCH 25/26] dmaengine: uniphier-xdmac: " Pan Chuang
2026-07-09 14:41   ` sashiko-bot
2026-07-09 13:58 ` [PATCH 26/26] dmaengine: xgene-dma: " Pan Chuang
2026-07-09 14:39   ` sashiko-bot
2026-07-09 14:26 ` [PATCH 00/26] dmaengine: Remove redundant error messages on IRQ request failure Wolfram Sang

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=20260709143002.B89541F000E9@smtp.kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=Frank.Li@kernel.org \
    --cc=dmaengine@vger.kernel.org \
    --cc=panchuang@vivo.com \
    --cc=sashiko-reviews@lists.linux.dev \
    --cc=vkoul@kernel.org \
    /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