Thread (25 messages) flat view 25 messages, 4 authors, 11d ago

Re: [PATCH PRELIMINARY 1/5] arm64: dts: qcom: shikra-iqs-som: fix GPIO138 reservation

From: Mohd Ayaan Anwar <mohd.anwar@oss.qualcomm.com>
Date: 2026-09-08 13:43:33
Also in: linux-arm-msm, linux-devicetree, lkml

On Tue, Sep 08, 2026 at 02:11:22PM +0200, Andrew Lunn wrote:
On Tue, Sep 08, 2026 at 11:53:30AM +0530, Mohd Ayaan Anwar wrote:
quoted
The GPIO mappings on the IQS variant differ from the CQ variants.
GPIO138 is connected to the RGMII1_RX_CTL pin rather than the NFC ESE
Secure IO pin; the latter is connected to GPIO49. This incorrect
reservation causes the probe of the second Ethernet port to fail:

  shikra-tlmm 500000.pinctrl: error -EINVAL: pin-138 (5d20000.ethernet)
  shikra-tlmm 500000.pinctrl: error -EINVAL: could not request pin 138
  (GPIO_138) from group gpio138 on device 500000.pinctrl
  qcom-ethqos 5d20000.ethernet: Error applying setting, reverse things back

Replace gpio138 with gpio49 in the reserved list.

Fixes: 779aead2dace ("arm64: dts: qcom: shikra: Add gpio-reserved-ranges to tlmm")
Signed-off-by: Mohd Ayaan Anwar <mohd.anwar@oss.qualcomm.com>
If this is a fix, should it be posted for a different tree, so it
makes it way to stable?

You don't normally mix fixes and on going development work.
You're right that it reads like a standalone fix, and that's on me for
the commit message. There's no functional impact on current users: the
reserved GPIO (138) is unused today, so applying this on a released
kernel produces no error and there's nothing to backport to stable.

I'll reword the commit message so it reads as a prerequisite for
enabling Ethernet rather than implying a current break.

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