Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Message-ID: <0dcd4c655bd9477d7adf2c6e32a1f0ebab18cee0.camel@baylibre.com>
Subject: Re: [PATCH] arm64: dts: meson: fix g12a buses
From:   Jerome Brunet <jbrunet@baylibre.com>
To:     Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Cc:     Kevin Hilman <khilman@baylibre.com>,
        linux-amlogic@lists.infradead.org, devicetree@vger.kernel.org,
        linux-kernel@vger.kernel.org, jianxin.pan@amlogic.com
Date:   Thu, 17 Jan 2019 22:45:18 +0100
In-Reply-To: <CAFBinCAdJu2C-LWspiHR4vVr=PUD2HGNi4zz+CSJP96ZE4GoJw@mail.gmail.com>
References: <20190116165236.8330-1-jbrunet@baylibre.com>
         <CAFBinCBFge2tir-MXw+ojTxzW+7n9Uf4FzCahrw0MDzAYV2SSg@mail.gmail.com>
         <f7289814e316ad5879848fb8c379db396f615828.camel@baylibre.com>
         <CAFBinCDhryOWHmRKp7GhWp2O+NpJziEgHoUE4HKVitTPDMrrBQ@mail.gmail.com>
         <ba1f56258e7873041a0f75996e376170563c9e6e.camel@baylibre.com>
         <CAFBinCAdJu2C-LWspiHR4vVr=PUD2HGNi4zz+CSJP96ZE4GoJw@mail.gmail.com>
Content-Type: text/plain; charset="UTF-8"
User-Agent: Evolution 3.30.4 (3.30.4-1.fc29) 
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Thu, 2019-01-17 at 22:20 +0100, Martin Blumenstingl wrote:
> On Thu, Jan 17, 2019 at 9:39 PM Jerome Brunet <jbrunet@baylibre.com> wrote:
> > On Thu, 2019-01-17 at 21:27 +0100, Martin Blumenstingl wrote:
> > > OK, but we had incorrect documentation in the past. did you check this
> > > with someone from Amlogic?
> > > 
> > > I'm curious because there seem to be two different approaches here:
> > > 1) hiubus name and offsets are being fixed within this patch
> > > 2) aobus is being dropped here and re-introduced with a different name
> > > later
> > > on
> > > 
> > 
> > because hiu exist and aobus does not, for which both the name and size was
> > wrong
> > 
> > > approach 1) can also be used for the "rti" region (at least in my
> > > opinion, the patch doesn't explain why it can't be done):
> > 
> > THe patch remove aobus (instead of fixing name and size) because, of the
> > multiple region documented covered by this 'made region', I did not
> > anticipate
> > which one will be required and I did not want to add them all.
> > 
> > Better to add them as needed, which is want I done for pinctrl as you
> > pointed
> > out
> > 
> > > rename "aobus" to "rti" and change the size to either 0x1000 or 0xb000
> > > (both values can be found in mesong12a.dtsi from
> > > buildroot_openlinux_kernel_4.9_fbdev_20180706)
> > 
> > RTI is added here:
> > https://lkml.kernel.org/r/20190117103151.3349-1-jbrunet@baylibre.com
> > 
> > I don't really understand the problem ? result is the same
> the actual problem is "me" as I have conflicting information:
> - Amlogic's buildroot kernel (for G12A) uses similar bus definitions
> as the GX SoCs (for which there are public datasheets) - this is how
> Jianxin added it to meson-g12a.dtsi originally

And it was the same for the GX family. AOBUS in the DT while there nothing
about this in the memory map. Keep wrong patterns does not make them right.

I'm merely reading the memory map here

> - this patch does it different - but cannot check if this is correct
> (no public datasheet is available for G12A or AXG) nor do I have a
> "big picture" of upcoming changes

Yes it does it differently. We should have picked up on this a while ago,
since gxbb at least, and we did not. There reason to create bus that don't
exist in the memory map of any recent SoC.

Creating bus should at least require a start and end offset explained
somewhere. Copying vendor DT is not enough

> 
> Cc'ing Jianxin: can you please review Jerome's patch and give some
> more details on the memory map on G12A so further contributions can be
> reviewed easier?
> 
> 
> Regards
> Martin