[RFC PATCH 0/4] Remove some e300/MPC83xx evaluation platforms

Li Yang leoyang.li at nxp.com
Tue Feb 28 07:42:58 AEDT 2023


On Sat, Feb 25, 2023 at 4:51 PM Arnd Bergmann <arnd at arndb.de> wrote:
>
> On Sat, Feb 25, 2023, at 17:50, Paul Gortmaker wrote:
> > [RE: [RFC PATCH 0/4] Remove some e300/MPC83xx evaluation platforms] On
> > 24/02/2023 (Fri 21:16) Leo Li wrote:
> >
> > Thanks for confirming with your marketing team that they "do not
> > recommend any new design with these SoCs" -- it also confirms the
> > information I read on the web pages for the platforms.
> >
> > As those of us immersed in this world all know from the 101 basics of
> > Product Life Cycle lessons, it doesn't matter if it is a phone or a
> > set-top-box/PVR or whatever else kind of non-PC consumer device --
> > kernel uprevs never happen in that product space.
> >
> > So with the best interests of the mainline kernel in mind, I think we
> > are good to proceed with this for summer 2023.  And of course as I've
> > said many times before - the kernel is in git, so really you can't
> > delete anything anyway - it remains in history forever.
>
> Thanks for working on this, this is a good step towards removing
> the known unused code. One aspect I'd add from doing similar cleanups
> on arm32 is that I would prioritize removing evaluation platforms
> for SoCs that have no other supported boards, and then
> garbage-collecting the device drivers that become unused.
>
> I'm not sure where the RDB boards fit in that, in particular if
> an unmodified kernel would work on a machine that is derived from
> the reference platform, or if it really only works on the machine
> itself. On most arm platforms, we moved to having only per-soc
> "compatible" strings, but on Freescale ppc32 it appears that
> the kernel always matches a board specific string and requires
> patches in order to support anything else.

The RDB boards normally don't include the complex configuration
mechanisms as used on the MDS platforms to support different
configurations at the same time.  They were designed to mimic a real
product and should be much more likely to work unmodified or with some
minor tweaks.

Regards,
Leo


More information about the Linuxppc-dev mailing list