linux-kernel - Re: [PATCH v5 1/2] dt-bindings: i2c: Add support for ASPEED i2Cv2

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:   Wed, 22 Feb 2023 09:30:14 +0800
From:   Jeremy Kerr <jk@...econstruct.com.au>
To:     Ryan Chen <ryan_chen@...eedtech.com>,
        Rob Herring <robh+dt@...nel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
        Joel Stanley <joel@....id.au>,
        Andrew Jeffery <andrew@...id.au>,
        Philipp Zabel <p.zabel@...gutronix.de>,
        "openbmc@...ts.ozlabs.org" <openbmc@...ts.ozlabs.org>,
        "linux-arm-kernel@...ts.infradead.org" 
        <linux-arm-kernel@...ts.infradead.org>,
        "linux-aspeed@...ts.ozlabs.org" <linux-aspeed@...ts.ozlabs.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v5 1/2] dt-bindings: i2c: Add support for ASPEED i2Cv2

Hi Ryan,

> > On the other hand, if it's just because of OS behaviour, then this
> > doesn't belong
> > in the DT.
> > 
> > Maybe to help us understand: why would you ever *not* want DMA
> > mode?
> > Isn't that always preferable?
> In AST SOC i2c design is 16 i2c bus share one dma engine. 

Does this mean that only one i2c controller in the system can be
configured to use DMA? Or is it able to be shared between multiple
controllers?

> It can be switch setting by dts setting. Otherwise driver by default
> probe is DMA mode.

You've explained what the modes do, and how they're switched, and what
the default is. However this doesn't explain *why* someone would want
to choose a particular mode when creating a controller node.

Still with the question above: assuming there are no restrictions on
DMA usage, why wouldn't a driver implementation just enable it always?

Cheers,

Jeremy