linux-kernel - Re: [PATCH v5] dmaengine: dw-edma: fix MSI data programming for multi-IRQ case

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: <aYJr4x2uADwhIKYQ@lizhi-Precision-Tower-5810>
Date: Tue, 3 Feb 2026 16:42:59 -0500
From: Frank Li <Frank.li@....com>
To: Shi-Shenghui <ssh.mediatek@...il.com>
Cc: vkoul@...nel.org, manivannan.sadhasivam@...aro.org,
	dmaengine@...r.kernel.org, linux-kernel@...r.kernel.org,
	brody.shi@...emi.com, kevin.song@...emi.com,
	qixiang.zhong@...emi.com, tom.hu@...emi.com,
	richard.yang@...emi.com
Subject: Re: [PATCH v5] dmaengine: dw-edma: fix MSI data programming for
 multi-IRQ case

On Mon, Feb 02, 2026 at 01:53:44PM +0800, Shi-Shenghui wrote:
> From: Shenghui Shi <brody.shi@...emi.com>
>
> When using MSI (not MSI-X) with multiple IRQs, the MSI data value
> must be unique per vector to ensure correct interrupt delivery.
> Currently, the driver fails to increment the MSI data per vector,
> causing interrupts to be misrouted.
>
> Fix this by caching the base MSI data and adjusting each vector's
> data accordingly during IRQ setup.
>
> Fixes: e63d79d1ff04 ("dmaengine: dw-edma: Add Synopsys DesignWare eDMA IP core driver")
>
> Signed-off-by: Shenghui Shi <brody.shi@...emi.com>
> ---
>  drivers/dma/dw-edma/dw-edma-core.c | 9 ++++++++-
>  1 file changed, 8 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/dma/dw-edma/dw-edma-core.c b/drivers/dma/dw-edma/dw-edma-core.c
> index 8e5f7defa6b6..dccc686b7a3e 100644
> --- a/drivers/dma/dw-edma/dw-edma-core.c
> +++ b/drivers/dma/dw-edma/dw-edma-core.c
> @@ -844,6 +844,7 @@ static int dw_edma_irq_request(struct dw_edma *dw,
>  {
>  	struct dw_edma_chip *chip = dw->chip;
>  	struct device *dev = dw->chip->dev;
> +	struct msi_desc *msi_desc;
>  	u32 wr_mask = 1;
>  	u32 rd_mask = 1;
>  	int i, err = 0;
> @@ -895,9 +896,15 @@ static int dw_edma_irq_request(struct dw_edma *dw,
>  					  &dw->irq[i]);
>  			if (err)
>  				goto err_irq_free;
> +			msi_desc = irq_get_msi_desc(irq);
> +			if (msi_desc) {
> +				bool is_msi;
>
> -			if (irq_get_msi_desc(irq))
>  				get_cached_msi_msg(irq, &dw->irq[i].msi);
> +				is_msi = msi_desc && !msi_desc->pci.msi_attrib.is_msix;

AI:
	Problem: Inside the if (msi_desc) block, checking msi_desc && again
is redundant since we already know msi_desc is non-NULL.


After remove 'msi_desc' check, needn't is_msi

	if (!msi_desc->pci.msi_attrib.is_msix) of below check

Frank
> +				if (is_msi)
> +					dw->irq[i].msi.data = dw->irq[0].msi.data + i;
> +			}
>  		}
>
>  		dw->nr_irqs = i;
> --
> 2.49.0.windows.1
>