Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Tue, 2 Jun 2020 18:34:58 +0200
From:   Drew Fustini <drew@beagleboard.org>
To:     Tony Lindgren <tony@atomide.com>
Cc:     Grygorii Strashko <grygorii.strashko@ti.com>,
        linux-omap@vger.kernel.org, linux-kernel@vger.kernel.org,
        =?iso-8859-1?Q?Beno=EEt?= Cousson <bcousson@baylibre.com>,
        Rob Herring <robh+dt@kernel.org>, devicetree@vger.kernel.org,
        Santosh Shilimkar <ssantosh@kernel.org>,
        Suman Anna <s-anna@ti.com>,
        Haojian Zhuang <haojian.zhuang@linaro.org>,
        Linus Walleij <linus.walleij@linaro.org>,
        linux-gpio@vger.kernel.org, jkridner@beagleboard.org,
        robertcnelson@gmail.com
Subject: Re: [PATCH] ARM: dts: AM33xx-l4: add gpio-ranges
Message-ID: <20200602163458.GA847883@x1>
References: <20200602131428.GA496390@x1>
 <803e2d78-28ba-0816-dbb5-d441d7659a91@ti.com>
 <20200602135155.GE37466@atomide.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <20200602135155.GE37466@atomide.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Tue, Jun 02, 2020 at 06:51:55AM -0700, Tony Lindgren wrote:
> * Grygorii Strashko <grygorii.strashko@ti.com> [200602 13:44]:
> > 
> > 
> > On 02/06/2020 16:14, Drew Fustini wrote:
> > > Add gpio-ranges properties to the gpio controller nodes.
> > > 
> > > These gpio-ranges were created based on "Table 9-10. CONTROL_MODULE
> > > REGISTERS" in the  "AM335x Technical Reference Manual" [0] and "Table
> > > 4-2. Pin Attributes" in the "AM335x Sitara Processor datasheet" [1].
> > > A csv file with this data is available for reference [2].
> > 
> > It will be good if you can explain not only "what" is changed, but
> > also "why" it's needed in commit message.
> 
> Also, please check (again) that this is the same for all the am3
> variants. For omap3, we had different pad assignments even between
> SoC revisions. Different pad routings should be easy to deal with
> in the dts if needed though.
> 
> Regards,
> 
> Tony

It appears that the only usage of am33xx-l4.dtsi is for am335x for which
specific parts mentioned in those dtsi files are 3352, 3358, and 3359.

$ git grep am33xx-l4.dtsi 
arch/arm/boot/dts/am33xx.dtsi:#include "am33xx-l4.dtsi"
$ git grep -l '#include "am33xx.dtsi"' arch/ |wc -l
27
$ git grep -l '#include "am33xx.dtsi"' arch/ |grep -v am335x |wc -l
0

Also, it appears that the only AM33xx parts that actually exist are [0]:

AM3351, AM3352, AM3354, AM3356, AM3357, AM3358, AM3359

I clicked on the datasheet link for each product page and while the URL
has the specific part number in it [1], they all end up loading the
exact same PDF. The header states:

"AM3359, AM3358, AM3357, AM3356, AM3354, AM3352, AM3351
SPRS717L – OCTOBER 2011 – REVISED MARCH 2020"

Thus, I do believe all SoC's using am33xx-l4.dtsi would have the same
memory map for the pin control registers and the same relationshop from
pin to gpio line.  For example, GPMC_A0 has mode 7 and it is labeled
gpio1_16.  conf_gpmc_a0 is at offset 840h which makes it pin 16.

Maybe am33xx-l4.dtsi should have actually been named am335x-l4.dtsi?

Though I suppose there is no point in changing that now.

thanks,
drew

[0] http://www.ti.com/processors/sitara-arm/am335x-cortex-a8/overview.html
[1] https://www.ti.com/lit/ds/symlink/am3359.pdf