linux-kernel - Re: [PATCH 1/3] dt/bindings: arm-pl330: add description of arm,pl330-periph-burst

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]

Date:   Fri, 19 Aug 2016 08:15:41 +0530
From:   Vinod Koul <vinod.koul@...el.com>
To:     Shawn Lin <shawn.lin@...k-chips.com>
Cc:     Lars-Peter Clausen <lars@...afoo.de>,
        Rob Herring <robh+dt@...nel.org>,
        Huibin Hong <huibin.hong@...k-chips.com>,
        Xing Zheng <zhengxing@...k-chips.com>,
        devicetree@...r.kernel.org, dianders@...omium.org,
        briannorris@...omium.org, Caesar Wang <wxt@...k-chips.com>,
        dmaengine@...r.kernel.org, linux-kernel@...r.kernel.org,
        linux-arm-kernel@...ts.infradead.org,
        linux-rockchip@...ts.infradead.org
Subject: Re: [PATCH 1/3] dt/bindings: arm-pl330: add description of
 arm,pl330-periph-burst

On Wed, Aug 17, 2016 at 04:11:03PM +0800, Shawn Lin wrote:
> Hi, Vinod and Lars-Peter
> 
> Ping.. Any better idea to share :)
> 
> On 2016/8/9 17:12, Shawn Lin wrote:
> >Hi Lars-Peter,
> >
> >在 2016/8/9 16:39, Lars-Peter Clausen 写道:
> >>On 08/05/2016 09:25 AM, Shawn Lin wrote:
> >>>Hi Vinod,
> >>>
> >>>在 2016/8/5 11:34, Vinod Koul 写道:
> >>>>On Fri, Aug 05, 2016 at 10:53:20AM +0800, Shawn Lin wrote:
> >>>>>This patch adds the "arm,pl330-periph-burst" for arm-pl330 to
> >>>>>support busrt mode.
> >>>>
> >>>>why should this be DT property. Only reason I can think of if some hw
> >>>>versions support this and some won't.
> >>>
> >>>yes, if we want to support burst mode, both of the master(pl330) and
> >>>client(several peripherals) should implement it, otherwise it will
> >>>be broken when enabling.
> >>
> >>As you said, it is up to the consumer peripheral whether it supports
> >>BURST,
> >>SINGLE or both. So this is a per client property, but you specify this
> >>as a
> >>a global property on the producer side.
> >
> >Thanks for comment.
> >
> >yup, but what is the proper way to add it ? :)
> >
> >
> >a) If pl330 support BURST as well as all the peripherals, we could
> >enable it.
> >
> >b) If pl300 support BURST, but all the peripherals don't support it,
> >we could not enable it.
> >
> >c) If pl300 support BURST, but not all the peripherals support it,
> >we also could not enable it.
> >
> >the burst feature of peripheral IP may be vendor-specific, but the
> >common driver for this peripheral are used for many many vendors which
> >means we could not check all of this info. It's very likely to break
> >them... I couldn't figure out how many upstreamed peripheral drivers
> >who are using pl300 either.
> >
> >So this check should be done by all this vendors but we could make
> >sure we don't break them before they check a), b), c), right?

Since support for BURST needs to be from peripheral too, we should have
that as a property for peripheral not for controller.

The peripheral drivers can communicate the burst to be used to pl330
using src_maxburst/dst_maxburst in dma_slave_config. We can use this
value to indicate the DMA should be single (a value of 0) or burst with
"burst" value.

-- 
~Vinod