| From: |
| Werner Sembach <wse-AT-tuxedocomputers.com> |
| To: |
| W_Armin-AT-gmx.de, hansg-AT-kernel.org, ilpo.jarvinen-AT-linux.intel.com |
| Subject: |
| [RFC PATCH 0/2] platform/x86: uniwill-laptop: Introduce device descriptor system |
| Date: |
| Tue, 25 Nov 2025 14:49:10 +0100 |
| Message-ID: |
| <20251125135729.487837-1-wse@tuxedocomputers.com> |
| Cc: |
| platform-driver-x86-AT-vger.kernel.org, linux-kernel-AT-vger.kernel.org, Werner Sembach <wse-AT-tuxedocomputers.com> |
| Archive-link: |
| Article |
The first patch of this RFC is based on a PoC by Armin and extended the be able
to modify feature bits based on readouts from the EC Memory.
The 2nd patch is a new feature that requires such a modification.
Armin what do you think? Is this a clean approach? Technically the feature bits
can be modified arbitrarily, but it should only ever be accessed by the slightly
modified getter method, so the risk should be very minimal.
I wonder if the static device_descriptor variable can somehow be skipped, but it
doesn't bother to have it there either IMHO.
Armin Wolf (1):
platform/x86: uniwill-laptop: Introduce device descriptor system
Werner Sembach (1):
platform/x86/uniwill: Implement cTGP setting
drivers/platform/x86/uniwill/uniwill-acpi.c | 269 +++++++++++++++++---
1 file changed, 240 insertions(+), 29 deletions(-)
--
2.43.0
