On Tue, 21 Nov 2023, Takashi Iwai wrote: > On Tue, 21 Nov 2023 16:10:12 +0100, > Lee Jones wrote: > > > > On Tue, 21 Nov 2023 15:26:29 +0100, Takashi Iwai wrote: > > > This reverts commit c7d80059b086c4986cd994a1973ec7a5d75f8eea. > > > > > > The commit caused regressions to a few drivers (such as Logitech > > > keyboard or System76 ACPI) that already had "color" leds sysfs entry; > > > now they conflict with the commonly created sysfs of led class. > > > > > > For addressing the regression, we revert the commit, so that the > > > conflicting "color" sysfs entry is removed. > > > > > > [...] > > > > Applied, thanks! > > > > [1/1] Revert "leds: class: Store the color index in struct led_classdev" > > commit: 17e1ad377bb1d39260e0594ee284d8e03f64fdfb > > It turned out that drivers/leds/rgb/leds-group-multicolor.c > initializes the cdev->color, and this revert breaks it. Indeed: linux/leds-fixes/drivers/leds/rgb/leds-group-multicolor.c: In function ‘leds_gmc_probe’: linux/leds-fixes/drivers/leds/rgb/leds-group-multicolor.c:109:49: error: ‘struct led_classdev’ has no member named ‘color’ linux/leds-fixes/drivers/leds/rgb/leds-group-multicolor.c:120:13: error: ‘struct led_classdev’ has no member named ‘color’ > So please scratch my previous revert patch. > I'm going to submit a partial revert instead. Sure. -- Lee Jones [李琼斯]
