Re: [PATCH 2/3] arm64: dts: freescale: Add i.MX91 dtsi support

[Pengfei Li <pengfei.li_1@nxp.com>]

On Fri, Nov 08, 2024 at 03:07:50PM +0100, Alexander Stein wrote:
> Hi,
> 
> Am Freitag, 8. November 2024, 11:06:26 CET schrieb Pengfei Li:
> > On Thu, Nov 07, 2024 at 02:06:27PM +0100, Alexander Stein wrote:
> > > Am Donnerstag, 7. November 2024, 13:49:50 CET schrieb Alexander Stein:
> > > > > diff --git a/arch/arm64/boot/dts/freescale/imx91.dtsi b/arch/arm64/boot/dts/freescale/imx91.dtsi
> > > > > new file mode 100644
> > > > > index 000000000000..a9f4c1fe61cc
> > > > > --- /dev/null
> > > > > +++ b/arch/arm64/boot/dts/freescale/imx91.dtsi
> > > > > @@ -0,0 +1,66 @@
> > > > > +// SPDX-License-Identifier: (GPL-2.0+ OR MIT)
> > > > > +/*
> > > > > + * Copyright 2024 NXP
> > > > > + */
> > > > > +
> > > > > +#include "imx91-pinfunc.h"
> > > > > +#include "imx93.dtsi"
> > > > > +
> > > > > +&{/thermal-zones/cpu-thermal/cooling-maps/map0} {
> > > > > +	cooling-device =
> > > > > +		<&A55_0 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
> > > > > +};
> > > > > +
> > > > > +&clk {
> > > > > +	compatible = "fsl,imx91-ccm";
> > > > > +};
> > > > > +
> > > > > +&eqos {
> > > > > +	clocks = <&clk IMX91_CLK_ENET1_QOS_TSN_GATE>,
> > > > > +			<&clk IMX91_CLK_ENET1_QOS_TSN_GATE>,
> > > > > +			<&clk IMX91_CLK_ENET_TIMER>,
> > > > > +			<&clk IMX91_CLK_ENET1_QOS_TSN>,
> > > > > +			<&clk IMX91_CLK_ENET1_QOS_TSN_GATE>;
> > > > > +	assigned-clocks = <&clk IMX91_CLK_ENET_TIMER>,
> > > > > +				<&clk IMX91_CLK_ENET1_QOS_TSN>;
> > > > > +	assigned-clock-parents = <&clk IMX93_CLK_SYS_PLL_PFD1_DIV2>,
> > > > > +					<&clk IMX93_CLK_SYS_PLL_PFD0_DIV2>;
> > > > 
> > > > Is it just me or is the alignment of new lines not matching?
> > > > 
> > > > 
> > > > > +};
> > > > > +
> > > > > +&fec {
> > > > > +	clocks = <&clk IMX91_CLK_ENET2_REGULAR_GATE>,
> > > > > +			<&clk IMX91_CLK_ENET2_REGULAR_GATE>,
> > > > > +			<&clk IMX91_CLK_ENET_TIMER>,
> > > > > +			<&clk IMX91_CLK_ENET2_REGULAR>,
> > > > > +			<&clk IMX93_CLK_DUMMY>;
> > > > > +	assigned-clocks = <&clk IMX91_CLK_ENET_TIMER>,
> > > > > +				<&clk IMX91_CLK_ENET2_REGULAR>;
> > > > > +	assigned-clock-parents = <&clk IMX93_CLK_SYS_PLL_PFD1_DIV2>,
> > > > > +					<&clk IMX93_CLK_SYS_PLL_PFD0_DIV2>;
> > > > 
> > > > Here as well: Is it just me or is the alignment of new lines not matching?
> > > > 
> > > > > +	assigned-clock-rates = <100000000>, <250000000>;
> > > > > +};
> > > > > +
> > > > > +&i3c1 {
> > > > > +	clocks = <&clk IMX93_CLK_BUS_AON>,
> > > > > +			<&clk IMX93_CLK_I3C1_GATE>,
> > > > > +			<&clk IMX93_CLK_DUMMY>;
> > > > > +};
> > > > > +
> > > > > +&i3c2 {
> > > > > +	clocks = <&clk IMX93_CLK_BUS_WAKEUP>,
> > > > > +			<&clk IMX93_CLK_I3C2_GATE>,
> > > > > +			<&clk IMX93_CLK_DUMMY>;
> > > > > +};
> > > > > +
> > > > > +&tmu {
> > > > > +	status = "disabled";
> > > > 
> > > > Why does the TMU needs to be disabled instead of deleted?
> > > > 
> > > > > +};
> > > > > +
> > > > > +/* i.MX91 only has one A core */
> > > > > +/delete-node/ &A55_1;
> > > > > +
> > > > > +/* i.MX91 not has cm33 */
> > > > > +/delete-node/ &cm33;
> > > > > +
> > > > > +/* i.MX91 not has power-domain@44461800 */
> > > > > +/delete-node/ &mlmix;
> > > > > 
> > > > 
> > > > Shouldn't the following node also be removed?
> > > > * mipi_csi
> > > > * dsi
> > > > * lvds_bridge
> > > > * lcdif_to_dsi
> > > > * lcdif_to_ldb
> > > 
> > > Add mu1 and mu2 to that list.
> > 
> > Hi, i.MX91 also has mu1 and mu2. so there is no need to remove them here.
> 
> Maybe you have more recent information. The RM available to me doesn't
> say anything about messaging unit, neither in memory map nor as a separate
> section describing the hardware. Memory area is marked as reserved.
> 
> If there are actually mu1 and mu2 available, what is on the B-side? There
> is no Cortex-M33 after all.
> 

Hi Alexander,

I rechecked the RM and found that it did remove MU1 and MU2, I will remove these two nodes in the next version.
Thanks for your comments!

BR,
Pengfei Li