[yocto] new system booting directly to halt
Chris Hallinan
challinan at gmail.com
Fri Sep 30 11:00:17 PDT 2016
Nope, you are seeing exactly what I would have expected. You have verified
a lot of functionality. Your kernel boots, mounts a root file system, and
executes your init command. So that's a big step. Now you need to sort
out your sysV init stuff. You can actually do a lot from that simple
shell. See below.
init does _lots_ of stuff on first boot, including mounting /proc and
/sys. On modern systems the /dev directory is populated dynamically as
devices are discovered. This is done by udev on a sysV system. While I
haven't looked into the details in anything modern ;), there used to be
just a tiny subset of static /dev nodes before udev took over. Don't know
how it exactly works today. But it doesn't matter, when you get it
working, it will all "Just Work" ;)
To get proc and sys, do this:
# mount -t proc /proc /proc
# mount -t sysfs /sys /sys
(Note to the experts: I know one of those parameters is superfluous, but it
displays nicely on mount command so I do it that way.)
Now that you have /proc and /sys, you can start executing the same commands
your init system would have done. Typically that comes from /etc/inittab,
and usually involves running something like a rc script with runlevel set
to 'S' at first. Take a look at inittab to see what it's supposed to do.
All of your sysV init scripts (and the symlinks which define which scripts
get run upon entry/exit of a runlevel) should be found in /etc/init.d/* if
memory serves. Haven't played with a sysV system in a while ;)
Even just try running /sbin/init and see what happens. Obviously something
in your setup is messed up.
Good luck,
Chris
On Fri, Sep 30, 2016 at 12:16 PM, Maxwell Bottiger <
sleepylight at jive-turkey.net> wrote:
> Chris,
>
> Thanks for the suggestions. The system is based on SystemV and uses the
> old school init scripts.
>
> I rebuilt my kernel with init=/bin/sh and I can get into the system now!
> The thing is, I don't see anything in /proc, and only /dev/null exists in
> the /dev directory. I don't know if the lack of meta filesystems is due to
> me skipping an actual init, or if I've buggered up something in my kernel
> configuration.
>
> This continues to be the world's most secure box, but I feel like I'm
> making progress on that.
>
> -Max
>
> On Thu, Sep 29, 2016 at 4:40 PM, Chris Hallinan <challinan at gmail.com>
> wrote:
>
>> You could sell this as a very secure box ;)
>>
>> Kidding aside, it didn't "zip" through runlevel 5, in fact, you don't
>> know what runlevel it is actually in. 11 seconds after your file system is
>> mounted, a reboot is issued. From your trace, nothing running in userspace
>> displayed any messages to the console. You didn't mention what type of
>> init system you are using, systemd or sysV. It almost looks like a console
>> problem. But...i don't know why a reboot would be issued. That's not
>> normal behavior if you simply have a min-configured console. Maybe init
>> will do that under certain circumstances, I am not sure.
>>
>> Just for FYI, notice that your serial port is instantiating on ttyS0. (I
>> do see that agrees with your kernel command line so at least that much is
>> correct.)
>>
>> First thing I would try is to see if I can get anything on userspace.
>> Try using init= on your kernel command line to just spawn a shell.
>> (init=/bin/sh)
>>
>> Do you have a way to examine the flash after a reboot to see if a syslog
>> was populated on the file system and examine its contents? ie JTAG?
>>
>> Make sure your root file system is sane, ie mount it on a development box
>> and make sure it's contents make sense to you. ie for sysV /sbin/init
>> should point to init, and it should contain a /etc/inittab and that should
>> have sane contents, etc. For systemd, /sbin/init should point (possibly
>> through /etc/alternatives) to a systemd executable. You get the idea.
>>
>> Good luck,
>>
>> Chris
>>
>>
>> On Thu, Sep 29, 2016 at 3:18 PM, Maxwell Bottiger <
>> sleepylight at jive-turkey.net> wrote:
>>
>>> I've been working on setting up a poky distro on an older intel XScale
>>> system, and I finally got a working kernel and root filesystem. However,
>>> when I boot the box it zips through runlevel 5, then halts. It seems to me
>>> like maybe init isn't kicking off getty or some sort of interactive login.
>>> Has anyone else run into this problem? I'll post the boot output below.
>>>
>>> -Max
>>>
>>> Bootloader Initialization Successful.
>>> Copyright (c) 2010 InHand Electronics (www.inhand.com)
>>> FT4: Fingertip4 v.0A211004 (Oct 21 2010, 10:04:57)
>>> IBL: InHandBoot_4.9 BETA v.0A211004 (Oct 21 2010, 10:04:19)
>>>
>>> CPLD Version: 5.10
>>> DRAM Size: 128 MBytes
>>> SRAM Size: 240 KBytes
>>> IBC Start Page: /nor/512
>>>
>>> Platform Post-Init
>>> ac97
>>>
>>> Autorun '/cf/autoexec.brc'
>>> '/cf/autoexec.brc' not found
>>>
>>> Autorun '/sd/autoexec.brc'
>>> '/sd/autoexec.brc' not found
>>>
>>> Autorun '/nor/config.brc'
>>> >> bootlin
>>> Starting Linux
>>> Copying image (2689792 bytes) to RAM: 0xa0008000
>>>
>>> Copying file /nor/zImage to /ram/os
>>> Source File: 2689792 bytes
>>> Destination Space: 4194304 bytes
>>>
>>> Copying 2689792 bytes
>>> pct remaining time
>>> --- --------- -----
>>> 100% 0 0:00
>>> Copied 2689792B (3MB)
>>> Setting up Kernel tags at: 0xa0000100
>>> Kernel tags used 36 bytes
>>> Disabling MMU
>>> Jumping to Linux kernel @ 0xa0008000
>>> Uncompressing Linux... done, booting the kernel.
>>> [ 0.000000] Booting Linux on physical CPU 0x0
>>> [ 0.000000] Linux version 3.14.4 (max at tsunami.pacs.agpea.army.mil)
>>> (gcc version 4.9.2 (GCC) ) #1 PREEMPT Wed Sep 28 16:29:41 EDT 2016
>>> [ 0.000000] CPU: XScale-PXA270 [69054118] revision 8 (ARMv5TE),
>>> cr=0000397f
>>> [ 0.000000] CPU: VIVT data cache, VIVT instruction cache
>>> [ 0.000000] Machine: InHand Fingertip4 Development Platform (aka FT4)
>>> [ 0.000000] Memory policy: Data cache writeback
>>> [ 0.000000] BUG: mapping for 0x00000000 at 0xff000000 out of vmalloc
>>> space
>>> [ 0.000000] Run Mode clock: 208.00MHz (*16)
>>> [ 0.000000] Turbo Mode clock: 520.00MHz (*2.5, active)
>>> [ 0.000000] Memory clock: 208.00MHz (/2)
>>> [ 0.000000] System bus clock: 208.00MHz
>>> [ 0.000000] Built 1 zonelists in Zone order, mobility grouping on.
>>> Total pages: 16256
>>> [ 0.000000] Kernel command line: console=ttyS0,115200
>>> root=/dev/mtdblock0 rw rootfstype=jffs2 mem=64M
>>> earlyprintk=serial,ttyS0,115200
>>> [ 0.000000] PID hash table entries: 256 (order: -2, 1024 bytes)
>>> [ 0.000000] Dentry cache hash table entries: 8192 (order: 3, 32768
>>> bytes)
>>> [ 0.000000] Inode-cache hash table entries: 4096 (order: 2, 16384
>>> bytes)
>>> [ 0.000000] Memory: 59616K/65536K available (3732K kernel code, 184K
>>> rwdata, 1132K rodata, 139K init, 94K bss, 5920K reserved)
>>> [ 0.000000] Virtual kernel memory layout:
>>> [ 0.000000] vector : 0xffff0000 - 0xffff1000 ( 4 kB)
>>> [ 0.000000] fixmap : 0xfff00000 - 0xfffe0000 ( 896 kB)
>>> [ 0.000000] vmalloc : 0xc4800000 - 0xff000000 ( 936 MB)
>>> [ 0.000000] lowmem : 0xc0000000 - 0xc4000000 ( 64 MB)
>>> [ 0.000000] modules : 0xbf000000 - 0xc0000000 ( 16 MB)
>>> [ 0.000000] .text : 0xc0008000 - 0xc04c83fc (4865 kB)
>>> [ 0.000000] .init : 0xc04c9000 - 0xc04ebcd4 ( 140 kB)
>>> [ 0.000000] .data : 0xc04ec000 - 0xc051a180 ( 185 kB)
>>> [ 0.000000] .bss : 0xc051a18c - 0xc0531b90 ( 95 kB)
>>> [ 0.000000] Preemptible hierarchical RCU implementation.
>>> [ 0.000000] NR_IRQS:16 nr_irqs:304 304
>>> [ 0.000046] sched_clock: 32 bits at 3250kHz, resolution 307ns, wraps
>>> every 1321528397516ns
>>> [ 0.000691] Console: colour dummy device 80x30
>>> [ 0.000769] Calibrating delay loop... 518.55 BogoMIPS (lpj=2592768)
>>> [ 0.060347] pid_max: default: 32768 minimum: 301
>>> [ 0.060594] Security Framework initialized
>>> [ 0.060741] Mount-cache hash table entries: 1024 (order: 0, 4096
>>> bytes)
>>> [ 0.060774] Mountpoint-cache hash table entries: 1024 (order: 0, 4096
>>> bytes)
>>> [ 0.062150] CPU: Testing write buffer coherency: ok
>>> [ 0.063287] Setting up static identity map for 0xa038e6d8 - 0xa038e744
>>> [ 0.069844] NET: Registered protocol family 16
>>> [ 0.071694] DMA: preallocated 256 KiB pool for atomic coherent
>>> allocations
>>> [ 0.095946] Configure Fingertip4...
>>> [ 0.188992] bio: create slab <bio-0> at 0
>>> [ 0.197249] usbcore: registered new interface driver usbfs
>>> [ 0.198138] usbcore: registered new interface driver hub
>>> [ 0.199246] usbcore: registered new device driver usb
>>> [ 0.210323] cfg80211: Calling CRDA to update world regulatory domain
>>> [ 0.211077] Switched to clocksource oscr0
>>> [ 0.282781] NET: Registered protocol family 2
>>> [ 0.284302] TCP established hash table entries: 1024 (order: 0, 4096
>>> bytes)
>>> [ 0.284382] TCP bind hash table entries: 1024 (order: 0, 4096 bytes)
>>> [ 0.284426] TCP: Hash tables configured (established 1024 bind 1024)
>>> [ 0.284648] TCP: reno registered
>>> [ 0.284673] UDP hash table entries: 256 (order: 0, 4096 bytes)
>>> [ 0.284717] UDP-Lite hash table entries: 256 (order: 0, 4096 bytes)
>>> [ 0.285236] NET: Registered protocol family 1
>>> [ 0.286343] RPC: Registered named UNIX socket transport module.
>>> [ 0.286402] RPC: Registered udp transport module.
>>> [ 0.286416] RPC: Registered tcp transport module.
>>> [ 0.286431] RPC: Registered tcp NFSv4.1 backchannel transport module.
>>> [ 0.292910] futex hash table entries: 256 (order: -1, 3072 bytes)
>>> [ 0.295619] NFS: Registering the id_resolver key type
>>> [ 0.295843] Key type id_resolver registered
>>> [ 0.295859] Key type id_legacy registered
>>> [ 0.295901] Installing knfsd (copyright (C) 1996 okir at monad.swb.de).
>>> [ 0.296346] jffs2: version 2.2. (NAND) © 2001-2006 Red Hat, Inc.
>>> [ 0.297841] msgmni has been set to 116
>>> [ 0.300940] io scheduler noop registered
>>> [ 0.300991] io scheduler deadline registered
>>> [ 0.301491] io scheduler cfq registered (default)
>>> [ 0.754840] pxa2xx-uart.0: ttyS0 at MMIO 0x40100000 (irq = 22,
>>> base_baud = 928562) is a UART1
>>> [ 1.153128] console [ttyS0] enabled
>>> [ 1.157921] pxa2xx-uart.1: ttyS1 at MMIO 0x40200000 (irq = 21,
>>> base_baud = 928562) is a UART2
>>> [ 1.168016] pxa2xx-uart.2: ttyS2 at MMIO 0x40700000 (irq = 20,
>>> base_baud = 928562) is a UART3
>>> [ 1.197503] brd: module loaded
>>> [ 1.221064] loop: module loaded
>>> [ 1.225694] nbd: registered device at major 43
>>> [ 1.271030] Uniform Multi-Platform E-IDE driver
>>> [ 1.277643] ide-gd driver 1.18
>>> [ 1.286435] physmap platform flash device: 04000000 at 00000000
>>> [ 1.293684] physmap-flash.0: Found 1 x16 devices at 0x0 in 16-bit
>>> bank. Manufacturer ID 0x000089 Chip ID 0x00891c
>>> [ 1.304429] physmap-flash.0: Found 1 x16 devices at 0x2000000 in
>>> 16-bit bank
>>> [ 1.311720] Intel/Sharp Extended Query Table at 0x010A
>>> [ 1.316876] Intel/Sharp Extended Query Table at 0x010A
>>> [ 1.322184] Intel/Sharp Extended Query Table at 0x010A
>>> [ 1.327328] Intel/Sharp Extended Query Table at 0x010A
>>> [ 1.332518] Intel/Sharp Extended Query Table at 0x010A
>>> [ 1.337660] Using buffer write method
>>> [ 1.341323] Using auto-unlock on power-up/resume
>>> [ 1.345909] cfi_cmdset_0001: Erase suspend on write enabled
>>> [ 1.353683] Creating 1 MTD partitions on "physmap-flash.0":
>>> [ 1.359284] 0x000000900000-0x000004000000 : "Filesystem"
>>> [ 1.371691] Using configured DiskOnChip probe address 0x4000000
>>> [ 1.380537] mousedev: PS/2 mouse device common for all mice
>>> [ 1.386463] i2c /dev entries driver
>>> [ 1.393931] TCP: cubic registered
>>> [ 1.397286] NET: Registered protocol family 17
>>> [ 1.402235] NET: Registered protocol family 15
>>> [ 1.406830] lib80211: common routines for IEEE802.11 drivers
>>> [ 1.413102] Key type dns_resolver registered
>>> [ 1.417577] XScale iWMMXt coprocessor detected.
>>> [ 3.259509] VFS: Mounted root (jffs2 filesystem) on device 31:0.
>>> [ 3.266333] Freeing unused kernel memory: 136K (c04c9000 - c04eb000)
>>> [ 14.484953] reboot: System halted
>>>
>>>
>>> --
>>> _______________________________________________
>>> yocto mailing list
>>> yocto at yoctoproject.org
>>> https://lists.yoctoproject.org/listinfo/yocto
>>>
>>>
>>
>>
>> --
>> Life is like Linux - it never stands still.
>>
>>
>
--
Life is like Linux - it never stands still.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.yoctoproject.org/pipermail/yocto/attachments/20160930/35b13da9/attachment.html>
More information about the yocto
mailing list