linux-kernel - Re: [PATCH v3] arm64: new board

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-next>] [day] [month] [year] [list]

Date:   Thu, 11 Oct 2018 15:20:11 +0200
From:   Andreas Färber <afaerber@...e.de>
To:     aleksandr.aleksandrov@...id.com,
        Maxime Ripard <maxime.ripard@...tlin.com>,
        Rob Herring <robh+dt@...nel.org>
Cc:     Mark Rutland <mark.rutland@....com>, Chen-Yu Tsai <wens@...e.org>,
        David Lechner <david@...hnology.com>,
        Thierry Reding <treding@...dia.com>,
        Noralf Trønnes <noralf@...nnes.org>,
        Johan Hovold <johan@...nel.org>,
        "devicetree@...r.kernel.org" <devicetree@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "linux-arm-kernel@...ts.infradead.org" 
        <linux-arm-kernel@...ts.infradead.org>
Subject: Re: [PATCH v3] arm64: new board - Emlid Neutis N5

Hi Aleksandr,

Please keep your replies in text-only format, not HTML.

Am 11.10.18 um 14:01 schrieb aleksandr.aleksandrov@...id.com:
>>      +/ {
>>      + model = "Emlid Neutis N5 Developer board";
>>      + compatible = "emlid,emlid-neutis-n5-devboard",
>>      + "emlid,emlid-neutis-n5",
>>
>> Do you need the two emlid there? What comes before the comma is the
>> vendor, while what is after is the model.
>  
> I think emlid-neutis-n5 module could be useful in the future, no need
> this now.

You misunderstand: The point would be to use, e.g., "emlid,neutis-n5"
instead of "emlid,emlid-neutis-n5" with duplicate "emlid,emlid-". It is
orthogonal to having multiple compatible strings.

>>  +&uart1 {
>>  + pinctrl-names = "default";
>>  + pinctrl-0 = <&uart1_pins>, <&uart1_rts_cts_pins>;
>>  + status = "okay";
>>  +};
>>
>> I guess this is for bluetooth? Have you tested serdev drivers?
>>
> Yes, bluetooth is connected over uart1.
> You mean if I have tested bluetooth stack via serial device?

Not quite, we're missing a child node within uart1 for a serdev driver.
Is there no such driver yet for your Bluetooth chipset, or did you not
yet check?

> Bluez works stably with bcm43xx over uart 1500000 baud rate.
>  
>>
>> Also, I have a general comments, and it really depends on what your
>> intention about the board ecosystem is. Do you expect the SOM to be
>> swappable in multiple boards, or do you expect to send it as something
>> that is just fixed into a daughter board?
>>
>> In the former case, you probably want to use overlays instead. In the
>> latter, you're fine.
>>
> Right, we expect the SoM to be swappable. I agree, to use overlays is
> more convenient, but
> the devboard DT file will be a reference for the overlays and the future
> boards based on Neutis.

What about just keeping the common nodes enabled in a SoM .dts, so that
the average board doesn't need an Overlay for booting?

@Maxime/Rob, is it possible to merge .dtso files these days? If not,
could that be considered in the big dts Makefile refactoring? :)

Cheers,
Andreas

-- 
SUSE Linux GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)