On Fri, Nov 03, 2023 at 11:23:47AM -0500, Ryan Eatmon wrote:
>
>
> On 11/2/2023 8:08 PM, Denys Dmytriyenko wrote:
> >On Thu, Nov 02, 2023 at 11:48:28AM +0100, Massimiliano Minella wrote:
> >>From: Ryan Eatmon <reat...@ti.com>
> >>
> >>Move to setting the values for PREFERRED_PROVIDER using the
> >>?= default assignment so that we can override the setting if
> >>we would like to.
> >>
> >>Signed-off-by: Ryan Eatmon <reat...@ti.com>
> >>Signed-off-by: Massimiliano Minella <massimiliano.mine...@se.com>
> >>---
> >> meta-ti-bsp/conf/machine/include/k3r5.inc | 6 +++---
> >> 1 file changed, 3 insertions(+), 3 deletions(-)
> >>
> >>diff --git a/meta-ti-bsp/conf/machine/include/k3r5.inc
> >>b/meta-ti-bsp/conf/machine/include/k3r5.inc
> >>index c5c03cbf..184d3a09 100644
> >>--- a/meta-ti-bsp/conf/machine/include/k3r5.inc
> >>+++ b/meta-ti-bsp/conf/machine/include/k3r5.inc
> >>@@ -11,9 +11,9 @@ require conf/machine/include/arm/armv7a/tune-cortexa8.inc
> >> #
> >> https://git.ti.com/cgit/ti-u-boot/ti-u-boot/tree/doc/board/ti/j721e_evm.rst
> >> #
> >> https://git.ti.com/cgit/ti-u-boot/ti-u-boot/tree/doc/board/ti/am62x_sk.rst
> >> # https://git.ti.com/cgit/ti-u-boot/ti-u-boot/tree/doc/board/ti/k3.rst
> >>-PREFERRED_PROVIDER_virtual/kernel = "linux-dummy"
> >>-PREFERRED_PROVIDER_virtual/bootloader = "u-boot-ti-staging"
> >>-PREFERRED_PROVIDER_u-boot = "u-boot-ti-staging"
> >>+PREFERRED_PROVIDER_virtual/kernel ?= "linux-dummy"
> >>+PREFERRED_PROVIDER_virtual/bootloader ?= "u-boot-ti-staging"
> >>+PREFERRED_PROVIDER_u-boot ?= "u-boot-ti-staging"
> >
> >Being able to override u-boot for k3r5 makes sense.
> >
> >But allowing to change the kernel for k3r5 multiconfig that is now marked as
> >baremetal is rather dubious...
> >
> >What kind of use-case requires it?
>
> I think the the use case that I created the original patch for was
> the upstream testing. We need to be able to change the kernel and
> uboot versions for the testing not having the ?= was causing issues.
Please note this is specifically for k3r5, a baremetal config, where
kernel is "neutered" by setting it to "linux-dummy". There are tons of
implicit dependencies deep inside the OE on the kernel and all its pieces,
modules, packages, and supporting components, which linux-dummy cuts out,
making it a NOP.
There's no reason to ever change it even for upstream builds and testing.
Not allowing it to be changed is a safeguard of sorts.
And since it is now being requested for backporting by Massimiliano, hence
my question about the use-case...
> And since this patch is on master and thus will be coming to the
> upcoming scarthgap branch shortly, I see no reason to not take it
> for kirkstone.
>
> But if we want to rethink all of this, I'm not opposed.
>
> >
> >> SPL_SUFFIX = "bin"
> >> SPL_BINARY =
> >> "tiboot3-${SYSFW_SOC}-${SYSFW_SUFFIX}-${SYSFW_CONFIG}.${SPL_SUFFIX}"
> >>--
> >>2.42.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#17237):
https://lists.yoctoproject.org/g/meta-ti/message/17237
Mute This Topic: https://lists.yoctoproject.org/mt/102339031/21656
Group Owner: meta-ti+ow...@lists.yoctoproject.org
Unsubscribe:
https://lists.yoctoproject.org/g/meta-ti/leave/6695321/21656/1393940836/xyzzy
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
