These patches enable the pwm backlight for the rk3288-evb board. There were tested by watching the backlight grow from off to max with the following instructions: cd /sys/class/backlight/backlight*/ for i in $(seq 255); do echo $i > brightness; sleep .01; done The first patch enables the proper IP. I think it could land in Heiko's tree. The second patch switches PWM cells from 2 to 3 on rk3288. I think it could land in Thierry's tree. With the PWM subsystem as currently structured, I believe this will break backward compatibility. However the rk3288 PWM driver was added so recently (and the rk3288 is so new) that this seems OK. In the worst case if someone yells that they are broken, I believe we could fix this in Linux by saying that if a PWM driver specifies 3 cells and the DTS definition of the PWM says 2 cells that we'll just pretend that we have no flags. The 3rd and 4th patches are DTS ones. They could land in Heiko's tree after the second patch has landed. They are based atop his current WIP 3.18 dts tree. Note that instantiating the PWM backlight will cause the system to hang unless Heiko's (clk: rockchip: protect critical clocks from getting disabled) <patchwork.kernel.org/patch/4725391> is landed. There are no compile time or runtime dependencies between these patches except that patch #3 needs to come before patch #4. ...and of course the PWM won't work without all 4 patches. Changes in v2: - Check for failed ioremap() Doug Anderson (4): ARM: rockchip: rk3288: Switch to use the proper PWM IP pwm: rockchip: Allow polarity invert on rk3288 ARM: dts: Add main PWM info to rk3288 ARM: dts: Enable pwm backlight on rk3288-EVB .../devicetree/bindings/pwm/pwm-rockchip.txt | 4 +- arch/arm/boot/dts/rk3288-evb.dtsi | 53 +++++++++++++++++ arch/arm/boot/dts/rk3288.dtsi | 68 ++++++++++++++++++++++ arch/arm/mach-rockchip/rockchip.c | 24 ++++++++ drivers/pwm/pwm-rockchip.c | 32 +++++++++- 5 files changed, 177 insertions(+), 4 deletions(-) -- 2.1.0.rc2.206.gedb03e5 -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
