On 4/5/24 2:18 AM, Jayesh Choudhary wrote:
Hi,
On 04/04/24 20:59, Andrew Davis wrote:
On 4/4/24 4:00 AM, Jayesh Choudhary wrote:
Include the uboot device tree files needed to boot the board.
Co-developed-by: Vaishnav Achath vaishnav.a@ti.com Signed-off-by: Vaishnav Achath vaishnav.a@ti.com Signed-off-by: Jayesh Choudhary j-choudhary@ti.com
arch/arm/dts/Makefile | 2 + arch/arm/dts/k3-j722s-binman.dtsi | 171 ++ arch/arm/dts/k3-j722s-ddr-lp4-50-3733.dtsi | 2795 ++++++++++++++++++++ arch/arm/dts/k3-j722s-evm-u-boot.dtsi | 36 + arch/arm/dts/k3-j722s-r5-evm.dts | 82 + 5 files changed, 3086 insertions(+) create mode 100644 arch/arm/dts/k3-j722s-binman.dtsi create mode 100644 arch/arm/dts/k3-j722s-ddr-lp4-50-3733.dtsi create mode 100644 arch/arm/dts/k3-j722s-evm-u-boot.dtsi create mode 100644 arch/arm/dts/k3-j722s-r5-evm.dts
[...]
diff --git a/arch/arm/dts/k3-j722s-evm-u-boot.dtsi b/arch/arm/dts/k3-j722s-evm-u-boot.dtsi new file mode 100644 index 0000000000..056ef08455 --- /dev/null +++ b/arch/arm/dts/k3-j722s-evm-u-boot.dtsi @@ -0,0 +1,36 @@ +// SPDX-License-Identifier: GPL-2.0 +/*
- Common J722S EVM dts file for SPLs
- Copyright (C) 2024 Texas Instruments Incorporated - https://www.ti.com/
- */
+#include "k3-j722s-binman.dtsi"
+/ { + chosen { + stdout-path = "serial2:115200n8"; + tick-timer = &main_timer0; + }; +};
+&main_pktdma { + reg = <0x00 0x485c0000 0x00 0x000100>, + <0x00 0x4a800000 0x00 0x020000>, + <0x00 0x4aa00000 0x00 0x040000>, + <0x00 0x4b800000 0x00 0x400000>, + <0x00 0x485e0000 0x00 0x020000>, + <0x00 0x484a0000 0x00 0x004000>, + <0x00 0x484c0000 0x00 0x002000>, + <0x00 0x48430000 0x00 0x004000>; + reg-names = "gcfg", "rchanrt", "tchanrt", "ringrt", + "cfg", "tchan", "rchan", "rflow";
Is this needed? Do we still not have the correct regs in upstream kernel?
I missed this change when I was going through j722s dts changes.
u-boot dts currently does not have this fixed. Tag v6.9-rc1-dts from devicetree-rebasing would have this fix.
Will keep this in mind.
Also from TRM memory map, the range is off for 'tchanrt' Will send a fix patch for that after checking for all platforms to linux upstream if required.
+};
+&dmsc { + bootph-pre-ram;
+ k3_sysreset: sysreset-controller { + compatible = "ti,sci-sysreset"; + bootph-pre-ram;
This node won't be needed soon either[0]. Should mean an almost empty -u-boot.dtsi file, which should be the goal.
Okay. I will remove the node.
Should I mark [0] as dependency or is it okay without it. Impact would only be on U-Boot RESET I think. Base support would still be functional.
That is correct, if this goes in before [0] without this node the only thing that wouldn't work is reset, and it would start working again when [0] goes in. So no real dependency.
Andrew
Thanks, Jayesh
Andrew
[0] https://lore.kernel.org/all/20240402160908.508974-1-afd@ti.com/
+ };
[...]