linux-kernel - Re: [PATCH v2 3/3] misc: fastrpc: Update dma

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

Message-ID: <2a6319e5-15e5-4cb7-a2f3-7521383fc30e@kernel.org>
Date: Wed, 15 Oct 2025 10:25:55 +0100
From: Srinivas Kandagatla <srini@...nel.org>
To: Kumari Pallavi <kumari.pallavi@....qualcomm.com>,
 kpallavi@....qualcomm.com, srini@...nel.org, amahesh@....qualcomm.com,
 arnd@...db.de, gregkh@...uxfoundation.org
Cc: quic_bkumar@...cinc.com, ekansh.gupta@....qualcomm.com,
 linux-kernel@...r.kernel.org, quic_chennak@...cinc.com,
 dri-devel@...ts.freedesktop.org, linux-arm-msm@...r.kernel.org,
 jingyi.wang@....qualcomm.com, aiqun.yu@....qualcomm.com,
 ktadakam@....qualcomm.com
Subject: Re: [PATCH v2 3/3] misc: fastrpc: Update dma_mask for CDSP support on
 Kaanapali SoC



On 10/15/25 5:57 AM, Kumari Pallavi wrote:
> DSP currently supports 32-bit IOVA (32-bit PA + 4-bit SID) for
> both Q6 and user DMA (uDMA) access. This is being upgraded to
> 34-bit PA + 4-bit SID due to a hardware revision in CDSP for
> Kaanapali SoC, which expands the DMA addressable range.
> Update DMA mask configuration in the driver to support CDSP on
> Kaanapali SoC. Set the default `dma_mask` to 32-bit and update
> it to 34-bit based on CDSP and SoC-specific compatible string.
> 
> Signed-off-by: Kumari Pallavi <kumari.pallavi@....qualcomm.com>
> ---
>  drivers/misc/fastrpc.c | 9 +++++++--
>  1 file changed, 7 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c
> index 1a5d620b23f2..f2e5e53e9067 100644
> --- a/drivers/misc/fastrpc.c
> +++ b/drivers/misc/fastrpc.c
> @@ -267,6 +267,7 @@ struct fastrpc_session_ctx {
>  
>  struct fastrpc_soc_data {
>  	u32 sid_pos;
> +	u32 cdsp_dma_mask;
How about making this an actual dmamask ex:

	u64 cdsp_dma_mask == DMA_BIT_MASK(64),
	u64 dma_mask == DMA_BIT_MASK(32),

This will give more clear picture of what is going on,

BTW, these values are not set in the patch for some reason for both
default and soc specific soc data>
>  struct fastrpc_channel_ctx {
> @@ -2178,6 +2179,7 @@ static int fastrpc_cb_probe(struct platform_device *pdev)
>  	int i, sessions = 0;
>  	unsigned long flags;
>  	int rc;
> +	u32 dma_mask = 32;this should come from default soc_data, do not hardcode this here.
	u64 dma_mask = default_soc_data->dma_mask;>
>  	cctx = dev_get_drvdata(dev->parent);
>  	if (!cctx)
> @@ -2197,6 +2199,9 @@ static int fastrpc_cb_probe(struct platform_device *pdev)
>  	sess->dev = dev;
>  	dev_set_drvdata(dev, sess);> +	if (cctx->domain_id == CDSP_DOMAIN_ID)
> +		dma_mask = cctx->soc_data->cdsp_dma_mask;
> +

>  	if (of_property_read_u32(dev->of_node, "reg", &sess->sid))
>  		dev_info(dev, "FastRPC Session ID not specified in DT\n");
>  
> @@ -2211,9 +2216,9 @@ static int fastrpc_cb_probe(struct platform_device *pdev)
>  		}
>  	}
>  	spin_unlock_irqrestore(&cctx->lock, flags);
> -	rc = dma_set_mask(dev, DMA_BIT_MASK(32));
> +	rc = dma_set_mask(dev, DMA_BIT_MASK(dma_mask));
>  	if (rc) {
> -		dev_err(dev, "32-bit DMA enable failed\n");
> +		dev_err(dev, "%u-bit DMA enable failed\n", dma_mask);
>  		return rc;
>  	}
>