Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 5 Feb 2025 02:39:19 +0100
From:      Mario Marietto <marietto2008@gmail.com>
To:        Mark Millard <marklmi@yahoo.com>
Cc:        Marco Devesas Campos <devesas.campos@gmail.com>, freebsd-arm <freebsd-arm@freebsd.org>
Subject:   Re: Raimbow screen installing / booting FreeBSD-14.2-RELEASE-arm64-aarch64-RPI.img on the raspberry pi zero 2w
Message-ID:  <CA%2B1FSijy9OSoaMi%2BLfeiD4zqj=_0OkYcM3OrMEwGZqxrV7X4pg@mail.gmail.com>
In-Reply-To: <D1B3EA48-16B7-4E9A-940E-A88B0E75F1DA@yahoo.com>
References:  <CA%2B1FSig9coHK-k1dNaong=rOSGGNW170gf%2B-13S7J3syORU_DA@mail.gmail.com> <5E12B96C-20A2-4FFB-A31E-00B259450707@gmail.com> <CA%2B1FSih6jzcy7po3Ux_nSfN6UF%2BzwBfp-iM1vzVgpPObnN3ukA@mail.gmail.com> <CA%2B1FSih9TauSj4YteGij2BFO7C4irxDE=RFoBtnKK6-EMzwV5Q@mail.gmail.com> <FEDBC538-E7B2-49CA-AD27-3641A1BF6E56@gmail.com> <D1B3EA48-16B7-4E9A-940E-A88B0E75F1DA@yahoo.com>

next in thread | previous in thread | raw e-mail | index | archive | help
--000000000000da5202062d5b3383
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

----> A better alternative experiment would likely be to:

I did :

marietto# cp bcm2710-rpi-3-b.dtb bcm2710-rpi-zero-2-w.dtb

and the board booted wonderfully without one only problem.

I think that it's a good idea to add a note on the bug report that the
patches created don't work and that the fix,for the moment is :

marietto# cp bcm2710-rpi-3-b.dtb bcm2710-rpi-zero-2-w.dtb

Thanks.

On Wed, Feb 5, 2025 at 2:35=E2=80=AFAM Mark Millard <marklmi@yahoo.com> wro=
te:

> On Feb 4, 2025, at 16:37, Marco Devesas Campos <devesas.campos@gmail.com>
> wrote:
>
> > The problem isn=E2=80=99t the fix =E2=80=94 it=E2=80=99s that there is =
simply no appropriate
> device tree for the zero-2 that is bundled with the image so the board ha=
s
> no idea how to boot.
> >
> > So, while not perfect, until someone from the projects adds a zero 2
> dtb, I=E2=80=99m guessing copying one of the pi-3 images is your best bet=
 (I just
> tried it and it works =E2=80=94 well, boots, at least).
>
> A better alternative experiment would likely be to:
>
> ) Note the PORTVERSION used for sysutils/rpi-firmware :
>   PORTVERSION=3D    1.20230405.g20230405
>   (That encodes the date 2023-Apr-05, by the way.)
>
> ) Look for the history of bcm2710-rpi-zero-2-w.dtb to
>   find one released (or at lest tagged) that is as
>   similar of a vintage as is available. In this case
>   there is a match for the specific version:
>
>
> https://github.com/raspberrypi/firmware/blob/1.20230405/boot/bcm2710-rpi-=
zero-2-w.dtb
>
> However, because it is a match, installing
> sysutils/rpi-firmware and looking at what
> it installs it turns out happens to have:
>
> /usr/local/share/rpi-firmware/bcm2710-rpi-zero-2-w.dtb
>
> So that is the file one would likely want to copy over
> to the relevant msdosfs and test rpi-zero-2-w operation
> with.
>
> But, I'll note that that the .dtb (or the live adjustments
> that the RPi* firmware makes after loading the *.dtb)
> may have things that the FreeBSD kernel does not support
> or does not handle/avoid. A lack of any explicit development
> or testing for the rpi-zero-2-w is likely why
> bcm2710-rpi-zero-2-w.dtb is not on the release or stable
> snapshot media.
>
> So it still may simply not work.
>
> I'll note that the bcm271[012]-rpi-*.dtb files are for
> aarch64 (a.k.a. arm64): Tier 1 (at least for non-embedded)
>
> Also, bcm2709-rpi-*.dtb files are for armv7 (32-bit)
> instead: Tier 2.
>
> Finally, bcm2708-rpi-*.dtb files are for armv6 (also
> 32-bit): Tier 2/3/Unsupported for 13/14/15 .
>
> =3D=3D=3D
> Mark Millard
> marklmi at yahoo.com
>
>
>

--=20
Mario.

--000000000000da5202062d5b3383
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div><span class=3D"gmail-im">----&gt; </span>A better alt=
ernative experiment would likely be to:</div><div><br></div><div>I did :</d=
iv><div><br></div><div>marietto# cp bcm2710-rpi-3-b.dtb bcm2710-rpi-zero-2-=
w.dtb</div><div><br></div><div>and the board booted wonderfully without one=
 only problem.</div><div><br></div><div>I think that it&#39;s a good idea t=
o add a note on the bug report that the patches created don&#39;t work and =
that the fix,for the moment is :</div><div><br></div><div><div>marietto# cp=
 bcm2710-rpi-3-b.dtb bcm2710-rpi-zero-2-w.dtb</div></div><div><br></div><di=
v>Thanks.<br></div></div><br><div class=3D"gmail_quote gmail_quote_containe=
r"><div dir=3D"ltr" class=3D"gmail_attr">On Wed, Feb 5, 2025 at 2:35=E2=80=
=AFAM Mark Millard &lt;<a href=3D"mailto:marklmi@yahoo.com">marklmi@yahoo.c=
om</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" style=3D"margi=
n:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex=
">On Feb 4, 2025, at 16:37, Marco Devesas Campos &lt;<a href=3D"mailto:deve=
sas.campos@gmail.com" target=3D"_blank">devesas.campos@gmail.com</a>&gt; wr=
ote:<br>
<br>
&gt; The problem isn=E2=80=99t the fix =E2=80=94 it=E2=80=99s that there is=
 simply no appropriate device tree for the zero-2 that is bundled with the =
image so the board has no idea how to boot.<br>
&gt; <br>
&gt; So, while not perfect, until someone from the projects adds a zero 2 d=
tb, I=E2=80=99m guessing copying one of the pi-3 images is your best bet (I=
 just tried it and it works =E2=80=94 well, boots, at least).<br>
<br>
A better alternative experiment would likely be to:<br>
<br>
) Note the PORTVERSION used for sysutils/rpi-firmware :<br>
=C2=A0 PORTVERSION=3D=C2=A0 =C2=A0 1.20230405.g20230405<br>
=C2=A0 (That encodes the date 2023-Apr-05, by the way.)<br>
<br>
) Look for the history of bcm2710-rpi-zero-2-w.dtb to<br>
=C2=A0 find one released (or at lest tagged) that is as<br>
=C2=A0 similar of a vintage as is available. In this case<br>
=C2=A0 there is a match for the specific version:<br>
<br>
<a href=3D"https://github.com/raspberrypi/firmware/blob/1.20230405/boot/bcm=
2710-rpi-zero-2-w.dtb" rel=3D"noreferrer" target=3D"_blank">https://github.=
com/raspberrypi/firmware/blob/1.20230405/boot/bcm2710-rpi-zero-2-w.dtb</a><=
br>
<br>
However, because it is a match, installing <br>
sysutils/rpi-firmware and looking at what<br>
it installs it turns out happens to have:<br>
<br>
/usr/local/share/rpi-firmware/bcm2710-rpi-zero-2-w.dtb<br>
<br>
So that is the file one would likely want to copy over<br>
to the relevant msdosfs and test rpi-zero-2-w operation<br>
with.<br>
<br>
But, I&#39;ll note that that the .dtb (or the live adjustments<br>
that the RPi* firmware makes after loading the *.dtb)<br>
may have things that the FreeBSD kernel does not support<br>
or does not handle/avoid. A lack of any explicit development<br>
or testing for the rpi-zero-2-w is likely why<br>
bcm2710-rpi-zero-2-w.dtb is not on the release or stable<br>
snapshot media.<br>
<br>
So it still may simply not work.<br>
<br>
I&#39;ll note that the bcm271[012]-rpi-*.dtb files are for<br>
aarch64 (a.k.a. arm64): Tier 1 (at least for non-embedded)<br>
<br>
Also, bcm2709-rpi-*.dtb files are for armv7 (32-bit)<br>
instead: Tier 2.<br>
<br>
Finally, bcm2708-rpi-*.dtb files are for armv6 (also<br>
32-bit): Tier 2/3/Unsupported for 13/14/15 .<br>
<br>
=3D=3D=3D<br>
Mark Millard<br>
marklmi at <a href=3D"http://yahoo.com" rel=3D"noreferrer" target=3D"_blank=
">yahoo.com</a><br>
<br>
<br>
</blockquote></div><div><br clear=3D"all"></div><br><span class=3D"gmail_si=
gnature_prefix">-- </span><br><div dir=3D"ltr" class=3D"gmail_signature">Ma=
rio.<br></div>

--000000000000da5202062d5b3383--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CA%2B1FSijy9OSoaMi%2BLfeiD4zqj=_0OkYcM3OrMEwGZqxrV7X4pg>