public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Hongren Zheng <i@zenithal.me>
To: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Indan Zupancic <Indan.Zupancic@mep-info.com>,
	"Michael S. Tsirkin" <mst@redhat.com>,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2] UIO: Let uio_pdrv_genirq support cached memory
Date: Thu, 18 May 2023 12:07:59 +0800	[thread overview]
Message-ID: <ZGWkn89Xr1YAP9K4@Sun> (raw)
In-Reply-To: <20220505121518.46054-1-Indan.Zupancic@mep-info.com>

On Thu, May 05, 2022 at 02:15:19PM +0200, Indan Zupancic wrote:
> If 'cacheable' DT property exists, the memory mapping will be
> done with UIO_MEM_IOVA instead of the default UIO_MEM_PHYS.
> 
> Signed-off-by: Indan Zupancic <Indan.Zupancic@mep-info.com>
> ---
>  Documentation/driver-api/uio-howto.rst | 3 ++-
>  drivers/uio/uio_pdrv_genirq.c          | 8 ++++++--
>  2 files changed, 8 insertions(+), 3 deletions(-)

I assume this is fundamentally OK? What else should be done to get this merged?

I plan to add similar functionality for uio_pci_generic,
especially for the RedHat IVSHMEM pcie device where
bar2 is just normal memory and should be marked
as UIO_MEM_IOVA as this has performance implication,
is that acceptable?

Cc'ed uio_pci_generic maintainer here.

> 
> diff --git a/Documentation/driver-api/uio-howto.rst b/Documentation/driver-api/uio-howto.rst
> index 907ffa3b38f5..3cb48fc757d9 100644
> --- a/Documentation/driver-api/uio-howto.rst
> +++ b/Documentation/driver-api/uio-howto.rst
> @@ -413,7 +413,8 @@ probed with the ``"of_id"`` module parameter set to the ``"compatible"``
>  string of the node the driver is supposed to handle. By default, the
>  node's name (without the unit address) is exposed as name for the
>  UIO device in userspace. To set a custom name, a property named
> -``"linux,uio-name"`` may be specified in the DT node.
> +``"linux,uio-name"`` may be specified in the DT node. If a property
> +named ``cacheable`` exists, the memory will be mapped cacheable.
>  
>  Using uio_dmem_genirq for platform devices
>  ------------------------------------------
> diff --git a/drivers/uio/uio_pdrv_genirq.c b/drivers/uio/uio_pdrv_genirq.c
> index 63258b6accc4..269784ce0ba3 100644
> --- a/drivers/uio/uio_pdrv_genirq.c
> +++ b/drivers/uio/uio_pdrv_genirq.c
> @@ -115,6 +115,7 @@ static int uio_pdrv_genirq_probe(struct platform_device *pdev)
>  	struct uio_mem *uiomem;
>  	int ret = -EINVAL;
>  	int i;
> +	int memtype = UIO_MEM_PHYS;
>  
>  	if (node) {
>  		const char *name;
> @@ -132,7 +133,10 @@ static int uio_pdrv_genirq_probe(struct platform_device *pdev)
>  		else
>  			uioinfo->name = devm_kasprintf(&pdev->dev, GFP_KERNEL,
>  						       "%pOFn", node);
> -
> +		if (of_get_property(node, "cacheable", NULL)) {
> +			dev_info(&pdev->dev, "%s cacheable\n", uioinfo->name);
> +			memtype = UIO_MEM_IOVA;
> +		}
>  		uioinfo->version = "devicetree";
>  		/* Multiple IRQs are not supported */
>  	}
> @@ -204,7 +208,7 @@ static int uio_pdrv_genirq_probe(struct platform_device *pdev)
>  			break;
>  		}
>  
> -		uiomem->memtype = UIO_MEM_PHYS;
> +		uiomem->memtype = memtype;
>  		uiomem->addr = r->start & PAGE_MASK;
>  		uiomem->offs = r->start & ~PAGE_MASK;
>  		uiomem->size = (uiomem->offs + resource_size(r)
> -- 
> 2.30.2
> 
> 

      reply	other threads:[~2023-05-18  4:08 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-05-04 14:52 [PATCH] UIO: Let uio_pdrv_genirq support cached memory Indan Zupancic
2022-05-04 15:29 ` Greg KH
2022-05-05 12:15 ` [PATCH v2] " Indan Zupancic
2023-05-18  4:07   ` Hongren Zheng [this message]

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=ZGWkn89Xr1YAP9K4@Sun \
    --to=i@zenithal.me \
    --cc=Indan.Zupancic@mep-info.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mst@redhat.com \
    /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