[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <1621003477-11250-1-git-send-email-michal.vokac@ysoft.com>
Date: Fri, 14 May 2021 16:44:35 +0200
From: Michal Vokáč <michal.vokac@...ft.com>
To: Pavel Machek <pavel@....cz>,
Jacek Anaszewski <jacek.anaszewski@...il.com>,
Rob Herring <robh+dt@...nel.org>,
Shawn Guo <shawnguo@...nel.org>
Cc: Fabio Estevam <festevam@...il.com>, devicetree@...r.kernel.org,
Sascha Hauer <s.hauer@...gutronix.de>,
Pengutronix Kernel Team <kernel@...gutronix.de>,
NXP Linux Team <linux-imx@....com>,
linux-kernel@...r.kernel.org, linux-leds@...r.kernel.org,
Michal Vokáč <michal.vokac@...ft.com>
Subject: [RFC 0/2] Regression in lp55xx driver since multicolor framework was added?
Hi,
Linux v5.9 and newer fails on our platform that uses the lp5562 LED
controller with the following error:
lp5562: probe of 1-0030 failed with error -22
The problem exists since introduction of the multicolor framework to
the lp55xx driver. It is caused by the lp55xx_parse_logical_led function
here [1]. Even if the LED multicolor framework is not enabled the color
property is required. But it did not used to be.
There are at least two other in-tree platforms that use the same
controller and could suffer from this problem.
I wonder what is a proper fix here. Either fix the driver to not
require the color property (do not know ho to do so) or update the binding
and DTs to reflect the real state of the driver (in this series).
Thank you in advance for any comments,
Michal
[1] https://elixir.bootlin.com/linux/v5.13-rc1/source/drivers/leds/leds-lp55xx-common.c#L639
Michal Vokáč (2):
dt-bindings: leds: Add color as a required property for lp55xx
controller
ARM: dts: imx6dl-yapp4: Fix lp5562 driver probe
Documentation/devicetree/bindings/leds/leds-lp55xx.yaml | 10 ++++++++++
arch/arm/boot/dts/imx6dl-yapp4-common.dtsi | 11 ++++-------
2 files changed, 14 insertions(+), 7 deletions(-)
--
2.7.4
Powered by blists - more mailing lists