On Tue, May 9, 2023, at 08:38, Geert Uytterhoeven wrote:
On Mon, May 8, 2023 at 10:01 PM Arnd Bergmann <arnd@xxxxxxxx> wrote:
On Mon, May 8, 2023, at 17:36, Niklas Schnelle wrote:
I think the m68k/atari and mips/dec variants don't necessarily
qualify as PIO, those are really just pointer dereferences, and
they don't use the actual inb/outb functions.
On atari, it looks like HAS_IOPORT may be set if ATARI_ROM_ISA
is, but on dec it's never enabled.
Atari does not use RTC_DRV_CMOS, but still relies on generic RTC
instead.
Ah right, I now remember working on that code, so we're good on
m68k then. I think it should work for everyone using
depends on HAS_IOPORT || ARCH_DECSTATION
in that case, as that is the only exception.
Last time (in 2013?) I tried converting to RTC_DRV_CMOS by registering
an "rtc_cmos" platform device, I couldn't get it to work.
If you ever want to revisit this, I suspect the harder part here
is to detach arch/m68k/ from the RTC_DRV_GENERIC code first, pushing
the device registration into the individual machine specific time.c
code. It's probably not even worth trying to share the rtc-cmos
driver, but it might be useful to share the library code like
RTC_DRV_ALPHA does.
Arnd
