Thread (10 messages) 10 messages, 3 authors, 2024-12-22

Re: [PATCH v3 1/3] fbdev: Fix recursive dependencies wrt BACKLIGHT_CLASS_DEVICE

From: "Arnd Bergmann" <arnd@arndb.de>
Date: 2024-12-22 20:50:40
Also in: dri-devel, linux-fbdev, linux-staging

On Sun, Dec 22, 2024, at 21:15, Helge Deller wrote:
On 12/22/24 17:09, Thomas Zimmermann wrote:
quoted
quoted
quoted
+    depends on BACKLIGHT_CLASS_DEVICE=y || BACKLIGHT_CLASS_DEVICE=FB_NVIDIA
Seems wrong. BACKLIGHT_CLASS_DEVICE is of type tristate.
There are more of those, please check.
It's exactly correct. Linking would fail with FB_NVIDIA=y and BACKLIGHT=m. The above construct avoids this case. Please see Arnd's review comment at [1].
I'm not arguing against "depends on BACKLIGHT_CLASS_DEVICE=y".
It's the "BACKLIGHT_CLASS_DEVICE=FB_NVIDIA" which is wrong.
BACKLIGHT_CLASS_DEVICE is tristate, so either "y", "n" or "m", but 
never "FB_NVIDIA".
There are multiple ways to express this, but that line is a correct
way to allow all of

  BACKLIGHT_CLASS_DEVICE=y, FB_NVIDIA=m, FB_NVIDIA_BACKLIGHT=y
  BACKLIGHT_CLASS_DEVICE=y, FB_NVIDIA=y, FB_NVIDIA_BACKLIGHT=y
  BACKLIGHT_CLASS_DEVICE=m, FB_NVIDIA=m, FB_NVIDIA_BACKLIGHT=y

but disallow the broken

  BACKLIGHT_CLASS_DEVICE=m, FB_NVIDIA=y, FB_NVIDIA_BACKLIGHT=y

If you find this line too confusing, can you suggest a different
expression that has the same behavior?

       Arnd
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help