5.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Tony Lindgren <tony@xxxxxxxxxxx>
[ Upstream commit 9b6a51aab5f5f9f71d2fa16e8b4d530e1643dfcb ]
With subtle timings changes, we can now sometimes get an external abort on
non-linefetch error booting am3 devices at sysc_reset(). This is because
of a missing reset delay needed for the usb target module.
Looks like we never enabled the delay earlier for am3, although a similar
issue was seen earlier with a similar usb setup for dm814x as described in
commit ebf244148092 ("ARM: OMAP2+: Use srst_udelay for USB on dm814x").
Cc: stable@xxxxxxxxxxxxxxx
Fixes: 0782e8572ce4 ("ARM: dts: Probe am335x musb with ti-sysc")
Signed-off-by: Tony Lindgren <tony@xxxxxxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
arch/arm/boot/dts/am33xx.dtsi | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/arm/boot/dts/am33xx.dtsi b/arch/arm/boot/dts/am33xx.dtsi
index f53695ecbb782..7daf6ae2cd9af 100644
--- a/arch/arm/boot/dts/am33xx.dtsi
+++ b/arch/arm/boot/dts/am33xx.dtsi
@@ -349,6 +349,7 @@ usb: target-module@47400000 {
<SYSC_IDLE_NO>,
<SYSC_IDLE_SMART>,
<SYSC_IDLE_SMART_WKUP>;
+ ti,sysc-delay-us = <2>;
clocks = <&l3s_clkctrl AM3_L3S_USB_OTG_HS_CLKCTRL 0>;
clock-names = "fck";
#address-cells = <1>;
--
2.43.0
