devicetree.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Robin Murphy <robin.murphy-5wv7dgnIgG8@public.gmane.org>
To: Liu Gang <Gang.Liu-3arQi8VN3Tc@public.gmane.org>,
	devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org,
	shawnguo-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org,
	robh-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org
Cc: minghuan.lian-3arQi8VN3Tc@public.gmane.org,
	leoyang.li-3arQi8VN3Tc@public.gmane.org,
	mingkai.hu-3arQi8VN3Tc@public.gmane.org,
	scott.wood-3arQi8VN3Tc@public.gmane.org
Subject: Re: [PATCH] PCI: layerscape: Add 'dma-coherent' property
Date: Tue, 7 Jun 2016 11:03:05 +0100	[thread overview]
Message-ID: <57569BD9.2060701@arm.com> (raw)
In-Reply-To: <1465282546-28256-1-git-send-email-Gang.Liu-3arQi8VN3Tc@public.gmane.org>

On 07/06/16 07:55, Liu Gang wrote:
> Add 'dma-coherent' description for PCI nodes.
>
> The 'dma-coherent' indicates that the hardware IP block can ensure
> the coherency of the data transferred from/to the IP block. This
> can avoid the software cache flush/invalid actions, and improve
> the performance significantly.

Note that depending on the exact details it may actually be *necessary* 
for correctness - if the properties of the system are such that 
cacheable writes from the root complex might allocate directly into some 
level of cache without going to RAM, then having the CPU think the 
device is non-coherent and thus invalidate the cache before reading back 
from the buffer will result in data loss. But yeah, the performance is 
the most visible aspect ;)

Robin.

> The PCI IP block of ls1043a has this capability, so adding
> this feature to improve the PCI performance.
>
> Signed-off-by: Liu Gang <Gang.Liu-3arQi8VN3Tc@public.gmane.org>
> ---
>   Documentation/devicetree/bindings/pci/layerscape-pci.txt | 4 ++++
>   1 file changed, 4 insertions(+)
>
> diff --git a/Documentation/devicetree/bindings/pci/layerscape-pci.txt b/Documentation/devicetree/bindings/pci/layerscape-pci.txt
> index ef683b2..41e9f55 100644
> --- a/Documentation/devicetree/bindings/pci/layerscape-pci.txt
> +++ b/Documentation/devicetree/bindings/pci/layerscape-pci.txt
> @@ -24,6 +24,9 @@ Required properties:
>     The first entry must be a link to the SCFG device node
>     The second entry must be '0' or '1' based on physical PCIe controller index.
>     This is used to get SCFG PEXN registers
> +- dma-coherent: Indicates that the hardware IP block can ensure the coherency
> +  of the data transferred from/to the IP block. This can avoid the software
> +  cache flush/invalid actions, and improve the performance significantly.
>
>   Example:
>
> @@ -38,6 +41,7 @@ Example:
>   		#address-cells = <3>;
>   		#size-cells = <2>;
>   		device_type = "pci";
> +		dma-coherent;
>   		num-lanes = <4>;
>   		bus-range = <0x0 0xff>;
>   		ranges = <0x81000000 0x0 0x00000000 0x40 0x00010000 0x0 0x00010000   /* downstream I/O */
>

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

  parent reply	other threads:[~2016-06-07 10:03 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-06-07  6:55 [PATCH] PCI: layerscape: Add 'dma-coherent' property Liu Gang
2016-06-07  6:55 ` [PATCH] arm64: dts: ls1043a: Add 'dma-coherent' for ls1043a PCI nodes Liu Gang
2016-06-08 20:31   ` Scott Wood
2016-06-12  3:00     ` Gang Liu
2016-06-08 19:54 ` [PATCH] PCI: layerscape: Add 'dma-coherent' property Rob Herring
     [not found] ` <1465282546-28256-1-git-send-email-Gang.Liu-3arQi8VN3Tc@public.gmane.org>
2016-06-07 10:03   ` Robin Murphy [this message]
2016-06-16  0:43   ` Shawn Guo

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=57569BD9.2060701@arm.com \
    --to=robin.murphy-5wv7dgnigg8@public.gmane.org \
    --cc=Gang.Liu-3arQi8VN3Tc@public.gmane.org \
    --cc=devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=leoyang.li-3arQi8VN3Tc@public.gmane.org \
    --cc=linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org \
    --cc=minghuan.lian-3arQi8VN3Tc@public.gmane.org \
    --cc=mingkai.hu-3arQi8VN3Tc@public.gmane.org \
    --cc=robh-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
    --cc=scott.wood-3arQi8VN3Tc@public.gmane.org \
    --cc=shawnguo-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.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;
as well as URLs for NNTP newsgroup(s).