From: Trevor Woerner <twoerner@gmail.com>
To: Quentin Schulz <quentin.schulz@theobroma-systems.com>
Cc: anthony.t.davies@gmail.com, yocto@lists.yoctoproject.org
Subject: Re: [yocto] [meta-rockchip] [PATCH v3] rock-3a add
Date: Wed, 25 Oct 2023 12:20:45 -0400 [thread overview]
Message-ID: <20231025162045.GA22432@localhost> (raw)
In-Reply-To: <5bfb47c7-98df-4bfc-bc11-ffc67276fd0c@theobroma-systems.com>
On Wed 2023-10-25 @ 05:49:06 PM, Quentin Schulz wrote:
> Hi Trevor,
>
> On 10/25/23 16:42, Trevor Woerner wrote:
> > On Wed 2023-10-25 @ 04:22:17 PM, Quentin Schulz wrote:
> > > Hi Trevor,
> > >
> > > On 10/25/23 15:56, Trevor Woerner via lists.yoctoproject.org wrote:
> > > > Hi Anthony,
> > > >
> > > > Thanks for working on this.
> > > >
> > > > On Mon 2023-10-16 @ 09:43:45 PM, anthony.t.davies@gmail.com wrote:
> > > > > From: Anthony Davies <anthony.t.davies@gmail.com>
> > > > >
> > > > > Add support for the Radxa Rock 3A
> > > > > https://wiki.radxa.com/Rock3/3a
> > > > >
> > > > > The TF-A project does not currently have support for
> > > > > the rk3568. Therefore, for the time-being, the only way to supply a
> > > > > TPL/DDR-init for the rk3568 is to use the closed-source rkbin binaries
> > > > > from Rockchip. If/when TF-A adds support for the rk3588 we can investigate
> > > > > switching.
> > > > >
> > > > > recipes-bsp/rkbin/rockchip-rkbin_git.bb was modified to allow a machine
> > > > > override to allow both rk3568 and rk3588s to use differnet binary blobs
> > > > >
> > > > > Signed-off-by: Anthony Davies <anthony.t.davies@gmail.com>
> > > > > ---
> > > > > README | 1 +
> > > > > conf/machine/include/rk3568.inc | 17 +++++++++++++++++
> > > > > conf/machine/rock-3a.conf | 12 ++++++++++++
> > > > > recipes-bsp/rkbin/rockchip-rkbin_git.bb | 16 +++++++++++++++-
> > > > > recipes-bsp/u-boot/u-boot%.bbappend | 5 +++++
> > > > > 5 files changed, 50 insertions(+), 1 deletion(-)
> > > > > create mode 100644 conf/machine/include/rk3568.inc
> > > > > create mode 100644 conf/machine/rock-3a.conf
> > > > >
> > > > > diff --git a/README b/README
> > > > > index 8104474..3357b47 100644
> > > > > --- a/README
> > > > > +++ b/README
> > > > > @@ -32,6 +32,7 @@ Status of supported boards:
> > > > > nanopi-r4s
> > > > > rock-5b
> > > > > nanopi-r2s
> > > > > + rock-3a
> > > >
> > > > There's still an issue with indents/tabs.
> > > >
> > > > > builds:
> > > > > marsboard-rk3066
> > > > > radxarock
> > > > > diff --git a/conf/machine/include/rk3568.inc b/conf/machine/include/rk3568.inc
> > > > > new file mode 100644
> > > > > index 0000000..5382b58
> > > > > --- /dev/null
> > > > > +++ b/conf/machine/include/rk3568.inc
> > > > > @@ -0,0 +1,17 @@
> > > > > +MACHINEOVERRIDES =. "rk3568:"
> > > > > +DEFAULTTUNE ?= "cortexa55"
> > > > > +
> > > > > +require conf/machine/include/arm/armv8-2a/tune-cortexa55.inc
> > > > > +require conf/machine/include/rockchip-defaults.inc
> > > > > +require conf/machine/include/rockchip-wic.inc
> > > > > +
> > > > > +KBUILD_DEFCONFIG ?= "defconfig"
> > > > > +KERNEL_FEATURES:append:rk3568 = " bsp/rockchip/remove-non-rockchip-arch-arm64.scc"
> > > > > +KERNEL_CLASSES ??= "kernel-fitimage"
> > > > > +KERNEL_IMAGETYPE ??= "fitImage"
> > > > > +
> > > > > +PREFERRED_PROVIDER_trusted-firmware-a = "rockchip-rkbin"
> > > > > +PREFERRED_PROVIDER_optee-os = "rockchip-rkbin"
> > > > > +
> > > > > +UBOOT_SUFFIX ?= "itb"
> > > > > +UBOOT_ENTRYPOINT ?= "0x06000000"
> > > > > diff --git a/conf/machine/rock-3a.conf b/conf/machine/rock-3a.conf
> > > > > new file mode 100644
> > > > > index 0000000..2ed83a3
> > > > > --- /dev/null
> > > > > +++ b/conf/machine/rock-3a.conf
> > > > > @@ -0,0 +1,12 @@
> > > > > +#@TYPE: Machine
> > > > > +#@NAME: Radxa Rock 3a
> > > > > +#@DESCRIPTION: ROCK3 is a series of Rockchip RK3566/RK3568 based
> > > > > +#SBC(Single Board Computer) and Compute Module by Radxa.
> > > > > +#https://wiki.radxa.com/Rock3
> > > > > +
> > > > > +require conf/machine/include/rk3568.inc
> > > > > +
> > > > > +KERNEL_DEVICETREE = "rockchip/rk3568-rock-3a.dtb"
> > > > > +MACHINE_EXTRA_RRECOMMENDS += "kernel-modules"
> > > > > +
> > > > > +UBOOT_MACHINE = "rock-3a-rk3568_defconfig"
> > > > > diff --git a/recipes-bsp/rkbin/rockchip-rkbin_git.bb b/recipes-bsp/rkbin/rockchip-rkbin_git.bb
> > > > > index 7fefb01..ad5593c 100644
> > > > > --- a/recipes-bsp/rkbin/rockchip-rkbin_git.bb
> > > > > +++ b/recipes-bsp/rkbin/rockchip-rkbin_git.bb
> > > > > @@ -14,6 +14,7 @@ S = "${WORKDIR}/git"
> > > > > COMPATIBLE_MACHINE = ""
> > > > > COMPATIBLE_MACHINE:rk3588s = "rk3588s"
> > > > > +COMPATIBLE_MACHINE:rk3568 = "rk3568"
> > > > > PACKAGE_ARCH = "${MACHINE_ARCH}"
> > > > > @@ -25,7 +26,16 @@ do_install() {
> > > > > PACKAGES = "${PN}"
> > > > > ALLOW_EMPTY:${PN} = "1"
> > > > > -do_deploy() {
> > > > > +do_deploy:rk3568() {
> > > > > + # Prebuilt TF-A
> > > > > + install -m 644 ${S}/bin/rk35/rk3568_bl31_v*.elf ${DEPLOYDIR}/bl31-rk3568.elf
> > > > > + # Prebuilt OPTEE-OS
> > > > > + install -m 644 ${S}/bin/rk35/rk3568_bl32_v*.bin ${DEPLOYDIR}/tee-rk3568.bin
> > > > > + # Prebuilt U-Boot TPL (DDR init)
> > > > > + install -m 644 ${S}/bin/rk35/rk3568_ddr_1560MHz_v1.18.bin ${DEPLOYDIR}/ddr-rk3568.bin
> > > > > +}
> > > > > +
> > > > > +do_deploy:rk3588s() {
> > > > > # Prebuilt TF-A
> > > > > install -m 644 ${S}/bin/rk35/rk3588_bl31_v*.elf ${DEPLOYDIR}/bl31-rk3588.elf
> > > > > # Prebuilt OPTEE-OS
> > > > > @@ -34,4 +44,8 @@ do_deploy() {
> > > > > install -m 644 ${S}/bin/rk35/rk3588_ddr_lp4_2112MHz_lp5_2736MHz_v*.bin ${DEPLOYDIR}/ddr-rk3588.bin
> > > > > }
> > > > > +do_deploy() {
> > > > > + bbfatal "COMPATIBLE_MACHINE requires a corressponding do_deploy:<MACHINE>() override"
> > > > > +}
> > > > > +
> > > >
> > > > This is not related to adding support for rock3a.
> > > >
> > >
> > > In a way it is, but would benefit in being in a different commit I agree.
> > >
> > > So basically Anthony's issue (if I remember correctly) was that if you're
> > > not overriding the default do_deploy, the recipe will pass but install the
> > > wrong files with the wrong names, without telling you.
> > >
> > > So instead, the thought was to make do_deploy fail by default (with bbfatal)
> > > unless it is overridden (e.g. for rk3588s and rk3568) so that people adding
> > > support for new SoCs don't wonder what's missing.
> > >
> > > I could see a commit with a diff like:
> > > """
> > > -do_deploy() {
> > > +do_deploy:rk3588s() {
> > > # Prebuilt TF-A
> > > install -m 644 ${S}/bin/rk35/rk3588_bl31_v*.elf
> > > ${DEPLOYDIR}/bl31-rk3588.elf
> > > # Prebuilt OPTEE-OS
> > > @@ -34,4 +44,8 @@ do_deploy() {
> > > install -m 644 ${S}/bin/rk35/rk3588_ddr_lp4_2112MHz_lp5_2736MHz_v*.bin
> > > ${DEPLOYDIR}/ddr-rk3588.bin
> > > }
> > >
> > > +do_deploy() {
> > > + bbfatal "COMPATIBLE_MACHINE requires a corressponding
> > > do_deploy:<MACHINE>() override"
> > > +}
> > > +
> > > """
> > >
> > > with a proper explanation in the commit log, and then a commit that adds one
> > > for rk3568.
> >
> > ...and then another set for kirkstone, no doubt?
> >
>
> I don't have a personal need for supporting anything else other than RK3588
> on kirkstone, so I will not be pushing for it. Therefore, this is a decision
> for you to make as a maintainer if you actively want people to do it as part
> of follow-up of commits to master branch, do it yourself or wait for people
> to contribute backports to kirkstone whenever they need them.
Okay, sounds good. I won't backport until/unless I know someone is interested.
Also, rk3568 support is probably poor in kirkstone anyway (same as rk3588/s).
next prev parent reply other threads:[~2023-10-25 16:20 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-16 10:43 [meta-rockchip] [PATCH v3] rock-3a add anthony.t.davies
2023-10-25 13:56 ` Trevor Woerner
2023-10-25 14:22 ` [yocto] " Quentin Schulz
2023-10-25 14:42 ` Trevor Woerner
2023-10-25 15:49 ` Quentin Schulz
2023-10-25 16:20 ` Trevor Woerner [this message]
2023-10-26 4:17 ` Anthony Davies
2024-01-21 2:41 ` Trevor Woerner
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20231025162045.GA22432@localhost \
--to=twoerner@gmail.com \
--cc=anthony.t.davies@gmail.com \
--cc=quentin.schulz@theobroma-systems.com \
--cc=yocto@lists.yoctoproject.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox