From 6640928e307791addb2b83b4778833ca18d6419a Mon Sep 17 00:00:00 2001
From: mooleshacat <43627985+mooleshacat@users.noreply.github.com>
Date: Mon, 15 Jun 2026 19:32:55 -0400
Subject: [PATCH] brain +5 wrinkles from inline thinking breakthroughs

---
 .../arch/arm/boot/dts/qcom/qcom-ipq4019-tew-829dru.dts         | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/target/linux/ipq40xx/files-6.12/arch/arm/boot/dts/qcom/qcom-ipq4019-tew-829dru.dts b/target/linux/ipq40xx/files-6.12/arch/arm/boot/dts/qcom/qcom-ipq4019-tew-829dru.dts
index 4fd5ca88e4..f12abf4e16 100644
--- a/target/linux/ipq40xx/files-6.12/arch/arm/boot/dts/qcom/qcom-ipq4019-tew-829dru.dts
+++ b/target/linux/ipq40xx/files-6.12/arch/arm/boot/dts/qcom/qcom-ipq4019-tew-829dru.dts
@@ -580,4 +580,7 @@ if we do not add the proprietary LED controller (why would we?) we free up that
 IF the LED controler occupied c000000 that means the mfr put the QCA8337 somewhere else. Do we need to find the address, and redefine QCA8337 at that address so the hardware address matches the switch address in software?
 we could keep QCA8075 where it is, given mfr also places it at same hardware address 90000.
 
+- Keep mdio (QCA8075) at 90000 (hardware address matches upstream:mfr)
+- Redefine upstream switch@c000000 at the correct hardware address from mfr
+
 */
\ No newline at end of file