i.MX ProcessorsのトピックRe: CPU1: failed to come online

CPU1: failed to come online

raymondman — Wed, 28 Sep 2022 11:03:21 GMT

The custom board is using iMX8M Mini processor. We upgrade from Hardknott to Kirkstone and use mender. But after flashing the image, we have two problems.

1) the RAM size has been set to 1GB in imx8mm_evk.h. The devicetree also has the following setting.

memory@40000000 {
device_type = "memory";
reg = <0x0 0x40000000 0 0x40000000>;
};

But U-boot shows it as 2GiB as follow.

U-Boot 2022.04-lf_v2022.04+g1c881f4da8 (Jun 07 2022 - 05:16:02 +0000)

CPU: i.MX8MMQ rev1.0 1800 MHz (running at 1200 MHz)
CPU: Commercial temperature grade (0C to 95C) at 58C
Reset cause: POR
Model: NXP i.MX8MM EVK board
DRAM: 2 GiB

2) The log also shows that other 3 CPU cannot come online.

[ 0.002781] smp: Bringing up secondary CPUs ...
[ 5.091634] CPU1: failed to come online
[ 5.091642] CPU1: failed in unknown state : 0x0
[ 10.215331] CPU2: failed to come online
[ 10.215338] CPU2: failed in unknown state : 0x0
[ 15.339028] CPU3: failed to come online
[ 15.339035] CPU3: failed in unknown state : 0x0
[ 15.339079] smp: Brought up 1 node, 1 CPU

In the Hardknott version, all can work perfectly. Does anyone have any advice?

Many thanks!

Re: CPU1: failed to come online

Chenyi_Luo — Tue, 03 Jan 2023 08:49:54 GMT

Hello,

I use iMX8QXP, imx-yocto-L5.4.24_2.1.0, also meet the problem.

[ 0.047913] smp: Bringing up secondary CPUs ...
[ 5.092627] CPU1: failed to come online
[ 5.092636] CPU1: failed in unknown state : 0x0
[ 10.217413] CPU2: failed to come online
[ 10.217422] CPU2: failed in unknown state : 0x0
[ 15.342180] CPU3: failed to come online
[ 15.342189] CPU3: failed in unknown state : 0x0
[ 15.342237] smp: Brought up 1 node, 1 CPU
[ 15.342243] SMP: Total of 1 processors activated.

Could you tell me how you solved the problem?

Thank you.

Re: CPU1: failed to come online

dpog — Fri, 03 Feb 2023 10:14:55 GMT

@raymondman, did you find a solution to your problem? I'm also having the problem with the 1GB/2GB in U-Boot, I suspect optee to cause the problem at my end. I got my kernel running by adding a kernel-boot-param "mem=1024MB" to circumvent this, perhaps this might help you too.