Re: [PATCH 1/4] gpio: sa1100: register software node for GPIO controller

["Arnd Bergmann" <arnd@arndb.de>]

On Thu, Oct 29, 2026, at 05:23, Guenter Roeck wrote:
> On Mon, Jul 06, 2026 at 08:14:00PM -0700, Dmitry Torokhov wrote:
>> Define and register a static software node (sa1100_gpiochip_node) for the
>> SA-1100 GPIO controller during sa1100_init_gpio(). Assign its firmware node
>> directly to the GPIO chip prior to calling gpiochip_add_data().
>> 
>> This allows StrongARM SA-1100 board setup files to reference the SoC GPIO
>> controller in property entries when converting legacy platform data to
>> software nodes, resolving pin bindings directly via the attached firmware
>> node without relying on name matching.
>> 
>> Assisted-by: Antigravity:gemini-3.5-flash
>> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
>
> With this patch in the tree, the qemu "collie" emulation no longer works.
> There is no console output.
>
> Bisect log is attached for reference.
>
> I am not suggesting a revert if that is even possible, but maybe it is time
> to retire collie support if no one else reported this problem.

Thanks for the report. My current plan is to retire all of sa1100 by
merging my board file removal series for v7.5-rc1, so it should be gone
from linux-next soon but still be available for the LTS release (whether
that is 7.3 or 7.4).

Hopefully Dmitry can spot an obvious fix for the regression, otherwise
a revert is probably best so we can leave collie working in the LTS
kernel and not spend much work on that.

      Arnd