Hi Uli,
On Wed, Feb 10, 2021 at 10:17 AM Ulrich Hecht <uli@xxxxxxxx> wrote:
> > On 02/04/2021 2:54 PM Geert Uytterhoeven <geert+renesas@xxxxxxxxx> wrote:
> > Questions:
> > - Do we want to bring up the Cortex-A7 cores, too? We don't on R-Car
> > H2 neither.
>
> What still needs to be done for that?
Revert ee490bcc4f2d456c ("ARM: shmobile: Extend APMU code to allow
single cluster only")? Looks like Magnus only wanted to enable the boot
cluster, due to lack of big.LITTLE scheduling support?
> [I recall having issues bringing up the A7 cores on my old APE6 board (that has since failed and been returned to Renesas). The kernel failed to bring up the A7 cores if enabled at boot, but I was able to enable up to three (any three) of them one-by-one via sysfs. I'm not sure if that was specific to that board, though.]
Disabling the check makes the A7 cores start on my APE6-EVM.
Userland (Debian nfsroot) boot fails in the middle though, when systemd
fails to send messages to one of its brethren.
With a ramdisk, everything seems fine. I can even offline/online CPU
cores from sysfs at will, which failed before!
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
