Hello Geert,

On Sun, 2025-06-08 at 11:39 +0200, Geert Uytterhoeven wrote:

Hi Adrian,

On Sat, 7 Jun 2025 at 14:08, John Paul Adrian Glaubitz
[off-list ref] wrote:

quoted

On Sat, 2025-03-15 at 11:59 +0100, Arnd Bergmann wrote:

quoted

From: Arnd Bergmann <arnd@arndb.de>

The ioread/iowrite functions on sh only do memory mapped I/O like the
generic verion, and never map onto non-MMIO inb/outb variants, so they
just add complexity. In particular, the use of asm-generic/iomap.h
ties the declaration to the x86 implementation.

Remove the custom versions and use the architecture-independent fallback
code instead. Some of the calling conventions on sh are different here,
so fix that by adding 'volatile' keywords where required by the generic
implementation and change the cpg clock driver to no longer depend on
the interesting choice of return types for ioread8/ioread16/ioread32.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>

quoted

Those are quite a number of changes that I would like to test on real hardware
first before merging them into the kernel.

@Geert: Could you test it on your SH-7751 LANDISK board as well?

Already done for a while, as this patch is commit 2494fce26e434071 ("sh:
remove duplicate ioread/iowrite helpers") in v6.15-rc1 ;-)

Well, there is no Tested-By from either of us, so this isn't optimal.

I wished Arnd could have at least pinged me back regarding this. He knows I'm
actively maintaining arch/sh and I would like to properly test and review
such changes.

But I'm not doing this professionally, so I cannot be always there with 100%
capacity. Just pushing such changes in without any input from me defeats the
purpose of a maintainer.

Adrian

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer
`. `'   Physicist
  `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913