linux-kernel - Re: [PATCH 2/5] dma: add dma mask interface 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]

Date:   Mon, 7 Aug 2023 15:50:20 +0800
From:   Baolin Wang <baolin.wang@...ux.alibaba.com>
To:     Kaiwei Liu <kaiwei.liu@...soc.com>, Vinod Koul <vkoul@...nel.org>,
        Orson Zhai <orsonzhai@...il.com>,
        Chunyan Zhang <zhang.lyra@...il.com>
Cc:     dmaengine@...r.kernel.org, linux-kernel@...r.kernel.org,
        kaiwei liu <liukaiwei086@...il.com>,
        Wenming Wu <wenming.wu@...soc.com>
Subject: Re: [PATCH 2/5] dma: add dma mask interface in probe



On 8/7/2023 1:20 PM, Kaiwei Liu wrote:
> In the probe of DMA, the default addressing range is 32 bits,
> while the actual DMA hardware addressing range used is 36 bits.
> So add dma_set_mask_and_coherent function to match DMA
> addressing range.
> 
> Signed-off-by: Kaiwei Liu <kaiwei.liu@...soc.com>
> ---
>   drivers/dma/sprd-dma.c | 9 +++++++++
>   1 file changed, 9 insertions(+)
> 
> diff --git a/drivers/dma/sprd-dma.c b/drivers/dma/sprd-dma.c
> index 20c3cb1ef2f5..0e146550dfbb 100644
> --- a/drivers/dma/sprd-dma.c
> +++ b/drivers/dma/sprd-dma.c
> @@ -1115,6 +1115,15 @@ static int sprd_dma_probe(struct platform_device *pdev)
>   	u32 chn_count;
>   	int ret, i;
>   
> +	ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(36));
> +	if (ret) {
> +		ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
> +		if (ret) {
> +			dev_err(&pdev->dev, "dma_set_mask_and_coherent failed\n");

The error message is not readable, may be something like below?
"unable to set coherent mask to 32\n"

The changes look good to me. Chunyan and Orson, please double check the 
DMA mask?

> +			return ret;
> +		}
> +	}
> +
>   	/* Parse new and deprecated dma-channels properties */
>   	ret = device_property_read_u32(&pdev->dev, "dma-channels", &chn_count);
>   	if (ret)