Fedora 17 Dreamplug Install Problem

[efloehr]

Hi,

I am trying to install Fedora 17 on my Dreamplug, but partway through boot, I get a "BUG: soft lockup - CPU#0 stuck for 22s! [udevd:242]" and I'm at a loss for the next step to take to determine the cause and fix for this problem.

First, I downloaded the kirkwood image at http://archive.kernel.org/fedora-sec...17/Images/arm/, and verified the checksum.

Then, on an unmounted 16GB SD card, I "xzcat /misc/efloehr/dreamplug/Fedora-17-arm-kirkwood-mmcblk0.img.xz > /dev/sde", mounted and verified the two partitions.

Then, I booted the Dreamplug, loaded the uImage and uInitrd, set the bootargs and booted, like so:

Marvell>> fatload usb 1 0x6400000 uImage-kirkwood
reading uImage-kirkwood

3291896 bytes read
Marvell>> fatload usb 1 0x7400000 uInitrd-kirkwood
reading uInitrd-kirkwood

10112646 bytes read
Marvell>> set bootargs 'console=ttyS0,115200 root=/dev/sdb2 rootdelay=20'
Marvell>> bootm 0x6400000 0x7400000

This resulted in a successful load:

## Booting kernel from Legacy Image at 06400000 ...
   Image Name:   3.4.2-3.fc17.armv5tel.kirkwood
   Created:      2012-06-18   4:58:01 UTC
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    3291832 Bytes = 3.1 MiB
   Load Address: 00008000
   Entry Point:  00008000
   Verifying Checksum ... OK
## Loading init Ramdisk from Legacy Image at 07400000 ...
   Image Name:   initramfs
   Created:      2012-06-18   4:58:08 UTC
   Image Type:   ARM Linux RAMDisk Image (uncompressed)
   Data Size:    10112582 Bytes = 9.6 MiB
   Load Address: 00000000
   Entry Point:  00000000
   Verifying Checksum ... OK
   Loading Kernel Image ... OK
OK

Starting kernel ...

And a successful kernel boot:

Uncompressing Linux... done, booting the kernel.
[    0.000000] Booting Linux on physical CPU 0
[    0.000000] Initializing cgroup subsys cpuset
[    0.000000] Initializing cgroup subsys cpu
[    0.000000] Linux version 3.4.2-3.fc17.armv5tel.kirkwood (mockbuild@hsv-trimslice-8-v5tel.farm.hsv.redhat.com) (gcc version 4.7.0 20120507 (Red Hat 4.7.0-5) (GCC) ) #1 Tue Jun 12 11:34:2
[    0.000000] CPU: Feroceon 88FR131 [56251311] revision 1 (ARMv5TE), cr=00053977
[    0.000000] CPU: VIVT data cache, VIVT instruction cache
[    0.000000] Machine: Marvell GuruPlug Reference Board
[    0.000000] Memory policy: ECC disabled, Data cache writeback
[    0.000000] Built 1 zonelists in Zone order, mobility grouping on.  Total pages: 130048
[    0.000000] Kernel command line: console=ttyS0,115200 root=/dev/sdb2 rootdelay=20
[    0.000000] PID hash table entries: 2048 (order: 1, 8192 bytes)
[    0.000000] Dentry cache hash table entries: 65536 (order: 6, 262144 bytes)
[    0.000000] Inode-cache hash table entries: 32768 (order: 5, 131072 bytes)
[    0.000000] allocated 1048576 bytes of page_cgroup
[    0.000000] please try 'cgroup_disable=memory' option if you don't want memory cgroups
[    0.000000] Memory: 512MB = 512MB total
[    0.000000] Memory: 501352k/501352k available, 22936k reserved, 0K highmem
[    0.000000] Virtual kernel memory layout:
[    0.000000]     vector  : 0xffff0000 - 0xffff1000   (   4 kB)
[    0.000000]     fixmap  : 0xfff00000 - 0xfffe0000   ( 896 kB)
[    0.000000]     vmalloc : 0xe0800000 - 0xff000000   ( 488 MB)
[    0.000000]     lowmem  : 0xc0000000 - 0xe0000000   ( 512 MB)
[    0.000000]     pkmap   : 0xbfe00000 - 0xc0000000   (   2 MB)
[    0.000000]     modules : 0xbf000000 - 0xbfe00000   (  14 MB)
[    0.000000]       .text : 0xc0008000 - 0xc05db798   (5966 kB)
[    0.000000]       .init : 0xc05dc000 - 0xc0622000   ( 280 kB)
[    0.000000]       .data : 0xc0622000 - 0xc06778e0   ( 343 kB)
[    0.000000]        .bss : 0xc0677904 - 0xc0751e2c   ( 874 kB)
[    0.000000] SLUB: Genslabs=13, HWalign=32, Order=0-3, MinObjects=0, CPUs=1, Nodes=1

... lines removed due to 10K message size limit ...

[   13.184305] EXT4-fs (sdb2): mounted filesystem with ordered data mode. Opts: (null)
[   13.332729] dracut: Checking ext4: /dev/sdb2
[   13.338615] dracut: issuing e2fsck -a  /dev/sdb2
[   13.449216] dracut: rootfs: clean, 19221/82720 files, 143206/330527 blocks
[   13.460087] dracut: Remounting /dev/sdb2 with -o ro
[   13.511348] EXT4-fs (sdb2): mounted filesystem with ordered data mode. Opts: (null)
[   13.546457] dracut: Mounted root filesystem /dev/sdb2
[   13.725030] dracut: Switching root
[   14.050959] type=1404 audit(1348631309.810:2): enforcing=1 old_enforcing=0 auid=4294967295 ses=4294967295
[   16.473557] SELinux:  Permission ptrace_child in class process not defined in policy.
[   16.481943] SELinux: the above unknown classes and permissions will be allowed
[   16.503930] type=1403 audit(1348631312.260:3): policy loaded auid=4294967295 ses=4294967295

Welcome to Fedora 17 (Beefy Miracle)!

But only partway through the OS load before I received the "soft lockup" which just keeps repeating itself:

Started Replay Read-Ahead Data                                         [  OK  ]
Starting Collect Read-Ahead Data...                                            
[   17.598986] systemd-readahead-collect[202]: Failed to mark /: Invalid argument
Starting Media Directory...                                                    
Started Lock Directory                                                 [  OK  ]
Starting Runtime Directory...                                                  
Starting udev Coldplug all Devices...                                          
Starting udev Kernel Device Manager...                                         

... lines removed due to 10K message size limit ...

Started Configure read-only root support                               [  OK  ]
[   19.717715] mv643xx_eth smi: probed
[   19.864633] mv643xx_eth_port mv643xx_eth_port.0: eth0: port 0 with MAC address f0:ad:4e:01:45:b2
[   19.913984] cfg80211: Calling CRDA to update world regulatory domain
[   19.980206] Bluetooth: Core ver 2.16
[   19.983868] NET: Registered protocol family 31
[   19.988366] Bluetooth: HCI device and connection manager initialized
[   47.746996] BUG: soft lockup - CPU#0 stuck for 22s! [udevd:242]
[   47.752942] Modules linked in: bluetooth(+) cfg80211 orion_nand(+) marvell rfkill nand nand_ecc nand_ids mtd mv_cesa mv643xx_eth(+) usb_storage mmc_block mvsdio sata_mv mmc_core
[   47.768967] 
[   47.770460] Pid: 242, comm:                udevd
[   47.775093] CPU: 0    Not tainted  (3.4.2-3.fc17.armv5tel.kirkwood #1)
[   47.781673] PC is at nand_read_byte+0x14/0x18 [nand]
[   47.786675] LR is at nand_command+0x194/0x1d4 [nand]
[   47.791663] pc : [<bf0c3014>]    lr : [<bf0c72e0>]    psr: 60000013
[   47.791669] sp : ddc0dd98  ip : bf0c714c  fp : c7967a20
[   47.803205] r10: bf0ef190  r9 : 00000000  r8 : ffffffff
[   47.808456] r7 : 000000ff  r6 : ffffffff  r5 : c7967a20  r4 : c7967800
[   47.815013] r3 : e0e32000  r2 : 00000081  r1 : ffffffff  r0 : 00000000
[   47.821570] Flags: nZCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment user
[   47.828738] Control: 0005397f  Table: 0796c000  DAC: 00000015
[   47.834539] [<c000f1e4>] (unwind_backtrace+0x0/0x124) from [<c0077a18>] (watchdog_timer_fn+0xf0/0x144)
[   47.843894] [<c0077a18>] (watchdog_timer_fn+0xf0/0x144) from [<c00388fc>] (__run_hrtimer+0xb0/0x1d0)
[   47.853074] [<c00388fc>] (__run_hrtimer+0xb0/0x1d0) from [<c0039138>] (hrtimer_interrupt+0x104/0x248)
[   47.862348] [<c0039138>] (hrtimer_interrupt+0x104/0x248) from [<c001602c>] (orion_timer_interrupt+0x24/0x34)
[   47.872225] [<c001602c>] (orion_timer_interrupt+0x24/0x34) from [<c00781a4>] (handle_irq_event_percpu+0x38/0x240)
[   47.882539] [<c00781a4>] (handle_irq_event_percpu+0x38/0x240) from [<c00783dc>] (handle_irq_event+0x30/0x40)
[   47.892420] [<c00783dc>] (handle_irq_event+0x30/0x40) from [<c007aa1c>] (handle_level_irq+0xbc/0xd0)
[   47.901600] [<c007aa1c>] (handle_level_irq+0xbc/0xd0) from [<c0077bd0>] (generic_handle_irq+0x28/0x38)
[   47.910957] [<c0077bd0>] (generic_handle_irq+0x28/0x38) from [<c0009b84>] (handle_IRQ+0x68/0x8c)
[   47.919790] [<c0009b84>] (handle_IRQ+0x68/0x8c) from [<c042bcf4>] (__irq_svc+0x34/0x80)
[   47.927849] [<c042bcf4>] (__irq_svc+0x34/0x80) from [<bf0c3014>] (nand_read_byte+0x14/0x18 [nand])
[   47.936878] [<bf0c3014>] (nand_read_byte+0x14/0x18 [nand]) from [<bf0c72e0>] (nand_command+0x194/0x1d4 [nand])
[   47.946954] [<bf0c72e0>] (nand_command+0x194/0x1d4 [nand]) from [<bf0c5e60>] (nand_scan_ident+0x184/0xb14 [nand])
[   47.957288] [<bf0c5e60>] (nand_scan_ident+0x184/0xb14 [nand]) from [<bf0c6824>] (nand_scan+0x34/0x54 [nand])
[   47.967182] [<bf0c6824>] (nand_scan+0x34/0x54 [nand]) from [<bf0ef104>] (orion_nand_probe+0x104/0x190 [orion_nand])
[   47.977686] [<bf0ef104>] (orion_nand_probe+0x104/0x190 [orion_nand]) from [<c025eaf0>] (platform_drv_probe+0x1c/0x20)
[   47.988350] [<c025eaf0>] (platform_drv_probe+0x1c/0x20) from [<c025d8fc>] (driver_probe_device+0x148/0x354)
[   47.998141] [<c025d8fc>] (driver_probe_device+0x148/0x354) from [<c025db78>] (__driver_attach+0x70/0x94)
[   48.007671] [<c025db78>] (__driver_attach+0x70/0x94) from [<c025bdb8>] (bus_for_each_dev+0x50/0x88)
[   48.016766] [<c025bdb8>] (bus_for_each_dev+0x50/0x88) from [<c025ce48>] (bus_add_driver+0x108/0x278)
[   48.025949] [<c025ce48>] (bus_add_driver+0x108/0x278) from [<c025de40>] (driver_register+0xa4/0x128)
[   48.035132] [<c025de40>] (driver_register+0xa4/0x128) from [<c025ed18>] (platform_driver_probe+0x20/0x74)
[   48.044749] [<c025ed18>] (platform_driver_probe+0x20/0x74) from [<c0008574>] (do_one_initcall+0x98/0x170)
[   48.054368] [<c0008574>] (do_one_initcall+0x98/0x170) from [<c005972c>] (sys_init_module+0x17d4/0x1978)
[   48.063810] [<c005972c>] (sys_init_module+0x17d4/0x1978) from [<c0008c40>] (ret_fast_syscall+0x0/0x2c)

Thanks much for any next steps you can help with!
Eric

[matth45]

I don't have much to help you with yet, but maybe some consolation: I don't think you're doing anything wrong. I've been experiencing this for the couple weeks on my sheevaplug. Here's what I've figured out so far:

3.4.X kernels seem to have some timer bug that leads to the "soft lockup" we see. I get the same orion_timer_interrupt BUG also.

3.5.X kernels have a hard lockup whenever the rtc is started. You can see at least one other person ran into this:

http://rglinuxtech.com/2012/07/30/ke...the-dreamplug/

I just tried 3.6.0-0.rc6.git2.1.fc18.armv5tel.kirkwood from the fedora 18/rawhide repo. Although the rtc bug from the 3.5.X kernels seems to be fixed, its right back to the orion_timer_interrupt soft lockup.

I've contacted the fedora-arm list once about this and got nothing. I'll try again soon. Clearly something is wrong with the kirkwood build.

The other thing I've noticed is that when it crashes like this it sets the rtc clock to a strange value. I think the driver for the hardware clock has a problem.

Here's the output from my soft lockup:

[  708.481257] BUG: soft lockup - CPU#0 stuck for 22s! [displaysign.py:592]
[  708.487988] Modules linked in: lockd sunrpc mtdchar ofpart vfat fat orion_nand nand nand_ecc nand_ids mtd mv_cesa mv643xx_eth leds_gpio mmc_block mvsdio mmc_core
[  708.502638] irq event stamp: 88943052
[  708.506318] hardirqs last  enabled at (88943051): [<c04df070>] _raw_spin_unlock_irqrestore+0x44/0x64
[  708.515509] hardirqs last disabled at (88943052): [<c04df454>] __irq_svc+0x34/0x80
[  708.523121] softirqs last  enabled at (88940908): [<c00211d8>] __do_softirq+0x2b4/0x394
[  708.531169] softirqs last disabled at (88940901): [<c0021778>] irq_exit+0x5c/0xc0
[  708.538695]
[  708.540187] Pid: 592, comm:       displaysign.py
[  708.544829] CPU: 0    Tainted: G        W     (3.6.0-0.rc6.git2.1.fc18.armv5tel.kirkwood #1)
[  708.553305] PC is at _raw_spin_unlock_irqrestore+0x4c/0x64
[  708.558817] LR is at mark_held_locks+0xdc/0x110
[  708.563368] pc : [<c04df078>]    lr : [<c00668ac>]    psr: 20000013
[  708.563368] sp : def89cb8  ip : 00000003  fp : 00000000
[  708.574893] r10: 00000000  r9 : 18343000  r8 : 00000023
[  708.580144] r7 : c12b7860  r6 : 00000013  r5 : c08b7dc0  r4 : 00000013
[  708.586701] r3 : dd651680  r2 : def89cb8  r1 : dd651680  r0 : 00000001
[  708.593259] Flags: nzCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment user
[  708.600427] Control: 0005397f  Table: 1de08000  DAC: 00000015
[  708.606221] [<c000ee60>] (unwind_backtrace+0x0/0x124) from [<c008da7c>] (watchdog_timer_fn+0x108/0x168)
[  708.615664] [<c008da7c>] (watchdog_timer_fn+0x108/0x168) from [<c00420b4>] (__run_hrtimer+0x228/0x488)
[  708.625018] [<c00420b4>] (__run_hrtimer+0x228/0x488) from [<c0042b60>] (hrtimer_interrupt+0x10c/0x270)
[  708.634376] [<c0042b60>] (hrtimer_interrupt+0x10c/0x270) from [<c0014170>] (orion_timer_interrupt+0x24/0x34)
[  708.644256] [<c0014170>] (orion_timer_interrupt+0x24/0x34) from [<c008e2f4>] (handle_irq_event_percpu+0xec/0x3bc)
[  708.654571] [<c008e2f4>] (handle_irq_event_percpu+0xec/0x3bc) from [<c008e608>] (handle_irq_event+0x44/0x64)
[  708.664449] [<c008e608>] (handle_irq_event+0x44/0x64) from [<c0090dec>] (handle_level_irq+0xd8/0xf0)
[  680.674696] [<c0090dec>] (handle_level_irq+0xd8/0xf0) from [<c008dc4c>] (generic_handle_irq+0x28/0x38)
[  680.684053] [<c008dc4c>] (generic_handle_irq+0x28/0x38) from [<c0009ba4>] (handle_IRQ+0x68/0x8c)
[  680.692885] [<c0009ba4>] (handle_IRQ+0x68/0x8c) from [<c04df458>] (__irq_svc+0x38/0x80)
[  680.700934] [<c04df458>] (__irq_svc+0x38/0x80) from [<c04df078>] (_raw_spin_unlock_irqrestore+0x4c/0x64)
[  680.710466] [<c04df078>] (_raw_spin_unlock_irqrestore+0x4c/0x64) from [<c026b210>] (check_unmap+0x6f4/0x754)
[  680.720344] [<c026b210>] (check_unmap+0x6f4/0x754) from [<c026b33c>] (debug_dma_unmap_page+0x64/0x70)
[  680.729616] [<c026b33c>] (debug_dma_unmap_page+0x64/0x70) from [<c0290844>] (dma_async_memcpy_pg_to_pg+0x218/0x27c)
[  680.740105] [<c0290844>] (dma_async_memcpy_pg_to_pg+0x218/0x27c) from [<c0291f74>] (dma_memcpy_pg_to_iovec+0xf0/0x180)
[  680.750860] [<c0291f74>] (dma_memcpy_pg_to_iovec+0xf0/0x180) from [<c03ef078>] (dma_skb_copy_datagram_iovec+0x100/0x1d4)
[  680.761785] [<c03ef078>] (dma_skb_copy_datagram_iovec+0x100/0x1d4) from [<c041d534>] (tcp_recvmsg+0x64c/0xae8)
[  680.771839] [<c041d534>] (tcp_recvmsg+0x64c/0xae8) from [<c0442a84>] (inet_recvmsg+0x48/0x5c)
[  680.780411] [<c0442a84>] (inet_recvmsg+0x48/0x5c) from [<c03bfde8>] (sock_aio_read+0x100/0x120)
[  680.789156] [<c03bfde8>] (sock_aio_read+0x100/0x120) from [<c010f340>] (do_sync_read+0x98/0xd4)
[  680.797900] [<c010f340>] (do_sync_read+0x98/0xd4) from [<c010fd08>] (vfs_read+0xb4/0x184)
[  680.806122] [<c010fd08>] (vfs_read+0xb4/0x184) from [<c010fe14>] (sys_read+0x3c/0x70)
[  680.813996] [<c010fe14>] (sys_read+0x3c/0x70) from [<c0008c60>] (ret_fast_syscall+0x0/0x38)

---------- Post added at 10:22 PM ---------- Previous post was at 10:18 PM ----------

https://bugzilla.redhat.com/show_bug...iple&id=845143

Seems to be a bug filed about this issue. Probably commenting there with a "me too" couldn't hurt, just so they know its not hardware specific.

---------- Post added at 10:37 PM ---------- Previous post was at 10:22 PM ----------

http://www.mail-archive.com/ath9k-de.../msg06651.html

Perhaps its the network stack.

[efloehr]

Thanks, I'll keep plugging away.

Do you know of any dependencies with versions of U-Boot, or is it well beyond that point when it's hanging?

[matth45]

I don't think this is related to u-boot (though I could certainly be wrong). But based on that bugzilla entry and a bit of googling it seems like it may be related to the networking stack. You happen to be hitting the bug during boot, but for me I get it after the system has been running for a few minutes. Others seem to have hit it at other times for other reasons.

One thought I had was to try an older kernel. I haven't gotten around to it yet, but I saw there's a fedora 15 arm kernel rpm available in the f15 repos. I'm not sure the f15 kernel would be bootable with the current systemd stuff in f17. If I get a chance to try it tonight I'll let you know how it goes.

[matth45]

I tried this kernel, but it wouldn't boot: kernel-kirkwood-2.6.41.6-1.fc15.armv5tel.rpm

It crashed trying to load the network driver.

mv643xx_eth: MV-643xx 10/100/1000 ethernet driver version 1.4

I also tried the f15 kernel from the updates repo: kernel-kirkwood-2.6.42.9-1.fc15.armv5tel.rpm

It also failed loading the network driver. Debug message was this:

[   10.226382] alg: hash: Test 1 failed for mv-hmac-sha1
[   10.231537] 00000000: 06 71 4d 7c cc cc b5 cf 1b d6 c7 ab d0 25 c4 21
[   10.238005] 00000010: 66 0b 8e 70
[   10.249368] mv643xx_eth: MV-643xx 10/100/1000 ethernet driver version 1.4
[   10.279642] ------------[ cut here ]------------
[   10.284321] WARNING: at fs/sysfs/dir.c:481 sysfs_add_one+0x84/0xac()
[   10.290720] sysfs: cannot create duplicate filename '/class/mdio_bus/0'
[   10.297360] Modules linked in: orion_nand(+) nand nand_ecc nand_ids rtc_mv(+) mtd mv643xx_eth(+) mv_cesa mmc_block mvsdio mmc_core
[   10.309275] [<c0014200>] (unwind_backtrace+0x0/0x124) from [<c00286d4>] (warn_slowpath_common+0x54/0x6c)
[   10.318818] [<c00286d4>] (warn_slowpath_common+0x54/0x6c) from [<c0028784>] (warn_slowpath_fmt+0x34/0x44)
[   10.328454] [<c0028784>] (warn_slowpath_fmt+0x34/0x44) from [<c013768c>] (sysfs_add_one+0x84/0xac)
[   10.337481] [<c013768c>] (sysfs_add_one+0x84/0xac) from [<c0137ef8>] (sysfs_do_create_link+0xfc/0x1e8)
[   10.346858] [<c0137ef8>] (sysfs_do_create_link+0xfc/0x1e8) from [<c026d8d8>] (device_add+0x200/0x58c)
[   10.356144] [<c026d8d8>] (device_add+0x200/0x58c) from [<c02b4dec>] (mdiobus_register+0x88/0x168)
[   10.365094] [<c02b4dec>] (mdiobus_register+0x88/0x168) from [<bf0396cc>] (mv643xx_eth_shared_probe+0x148/0x3a0 [mv643xx_eth])
[   10.376490] [<bf0396cc>] (mv643xx_eth_shared_probe+0x148/0x3a0 [mv643xx_eth]) from [<c0271324>] (platform_drv_probe+0x1c/0x20)
[   10.387960] [<c0271324>] (platform_drv_probe+0x1c/0x20) from [<c0270160>] (driver_probe_device+0x158/0x2b8)
[   10.397768] [<c0270160>] (driver_probe_device+0x158/0x2b8) from [<c0270328>] (__driver_attach+0x68/0x8c)
[   10.407316] [<c0270328>] (__driver_attach+0x68/0x8c) from [<c026f0fc>] (bus_for_each_dev+0x54/0x8c)
[   10.416418] [<c026f0fc>] (bus_for_each_dev+0x54/0x8c) from [<c026f8b8>] (bus_add_driver+0xd8/0x25c)
[   10.425522] [<c026f8b8>] (bus_add_driver+0xd8/0x25c) from [<c02705f4>] (driver_register+0xa8/0x134)
[   10.434638] [<c02705f4>] (driver_register+0xa8/0x134) from [<bf04000c>] (mv643xx_eth_init_module+0xc/0x3c [mv643xx_eth])
[   10.445591] [<bf04000c>] (mv643xx_eth_init_module+0xc/0x3c [mv643xx_eth]) from [<c0008574>] (do_one_initcall+0x98/0x170)
[   10.456534] [<c0008574>] (do_one_initcall+0x98/0x170) from [<c005a288>] (sys_init_module+0x1710/0x18b4)
[   10.465996] [<c005a288>] (sys_init_module+0x1710/0x18b4) from [<c000dae0>] (ret_fast_syscall+0x0/0x2c)
[   10.475359] ---[ end trace b1a3bb64d0c5b5cd ]---
[   10.496257] mii_bus 0 failed to register
[   10.542961] mv643xx_eth: probe of mv643xx_eth.0 failed with error -12
[   10.586952] rtc-mv rtc-mv: rtc core: registered rtc-mv as rtc0
[   10.604600] NAND device: Manufacturer ID: 0xad, Chip ID: 0xdc (Hynix NAND 512MiB 3,3V 8-bit)
[   10.635262] Scanning device for bad blocks
[   11.076007] Creating 3 MTD partitions on "orion_nand":
[   11.081462] 0x000000000000-0x000000100000 : "u-boot"
[   11.090663] 0x000000100000-0x000000500000 : "uImage"
[   11.103420] 0x000000500000-0x000020000000 : "root"
[   11.133177] Unable to handle kernel NULL pointer dereference at virtual address 00000000
[   11.141363] pgd = de044000
[   11.144081] [00000000] *pgd=00000000
[   11.147677] Internal error: Oops: 5 [#1]
[   11.151615] Modules linked in: mtdchar ofpart orion_nand(+) nand nand_ecc nand_ids rtc_mv mtd mv643xx_eth(+) mv_cesa mmc_block mvsdio mmc_core
[   11.164567] CPU: 0    Tainted: G        W     (2.6.42.9-1.fc15.armv5tel.kirkwood #1)
[   11.172367] PC is at mv643xx_eth_probe+0x8c/0x574 [mv643xx_eth]
[   11.178321] LR is at mv643xx_eth_probe+0x84/0x574 [mv643xx_eth]
[   11.184267] pc : [<bf03b8cc>]    lr : [<bf03b8c4>]    psr: a0000013
[   11.184273] sp : de4b7e18  ip : 00000000  fp : 00000001
[   11.195809] r10: 0000001c  r9 : 00000024  r8 : c063bcd8
[   11.201060] r7 : ddd9c400  r6 : bf03dcf0  r5 : c0639470  r4 : ddd9c000
[   11.207616] r3 : c063bce0  r2 : 00000008  r1 : 00000000  r0 : 00000000
[   11.214175] Flags: NzCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment user
[   11.221342] Control: 0005397f  Table: 1e044000  DAC: 00000015
[   11.227114] Process udevd (pid: 203, stack limit = 0xde4b6270)
[   11.232974] Stack: (0xde4b7e18 to 0xde4b8000)
[   11.237353] 7e00:                                                       df80a298 c063bce0
[   11.245571] 7e20: df886140 00000000 bf03dcf0 c063bce0 bf03dcf0 bf03dcf0 00000000 00000001
[   11.253795] 7e40: 00000024 0000001c 00000001 c0271324 c0271308 c0270160 bf03dcf0 c063bce0
[   11.262019] 7e60: c063bce0 c063bd14 bf03dcf0 c0270328 bf03dcf0 de4b7e80 c02702c0 c026f0fc
[   11.270241] 7e80: df80a278 df87da50 bf03dcf0 bf03dcf0 ddd31300 c06619e0 00000000 c026f8b8
[   11.278463] 7ea0: bf03db6f bf03db70 0000006d bf03dcf0 bf040000 00000000 ddc0d640 00000001
[   11.286677] 7ec0: 0000001c c02705f4 c0271358 00000000 bf040000 00000000 ddc0d640 00000001
[   11.294901] 7ee0: 0000001c bf04001c bf03dd2c c0008574 00000000 00000001 bf03dd2c bf03dd2c
[   11.303124] 7f00: bf03dd74 bf03dd2c bf03dd74 00000001 ddc0d640 00000001 00000024 c005a288
[   11.311347] 7f20: bf03dd38 00000000 dde4f6e0 00000124 c005793c c0441634 bf03de98 400e5368
[   11.319569] 7f40: e0860000 000085dc e0867fec e0867e74 e0867274 dd972200 00004eb0 00005950
[   11.327791] 7f60: 00000000 00000000 00000022 00000023 0000001d 00000000 00000017 00000000
[   11.336007] 7f80: 00000075 00000000 00020000 40ba8ad8 00000080 c000dcc8 de4b6000 00000000
[   11.344230] 7fa0: 00000000 c000dae0 00000000 00020000 4016b000 000085dc 400e5368 40ba8ad8
[   11.352453] 7fc0: 00000000 00020000 40ba8ad8 00000080 40ba8a60 000085dc 400e5368 00000000
[   11.360676] 7fe0: 402ea140 be8f9d50 400dbcbc 402ea150 60000010 4016b000 e92d4000 ebfffffe
[   11.368925] [<bf03b8cc>] (mv643xx_eth_probe+0x8c/0x574 [mv643xx_eth]) from [<c0271324>] (platform_drv_probe+0x1c/0x20)
[   11.379681] [<c0271324>] (platform_drv_probe+0x1c/0x20) from [<c0270160>] (driver_probe_device+0x158/0x2b8)
[   11.389471] [<c0270160>] (driver_probe_device+0x158/0x2b8) from [<c0270328>] (__driver_attach+0x68/0x8c)
[   11.399002] [<c0270328>] (__driver_attach+0x68/0x8c) from [<c026f0fc>] (bus_for_each_dev+0x54/0x8c)
[   11.408095] [<c026f0fc>] (bus_for_each_dev+0x54/0x8c) from [<c026f8b8>] (bus_add_driver+0xd8/0x25c)
[   11.417192] [<c026f8b8>] (bus_add_driver+0xd8/0x25c) from [<c02705f4>] (driver_register+0xa8/0x134)
[   11.426297] [<c02705f4>] (driver_register+0xa8/0x134) from [<bf04001c>] (mv643xx_eth_init_module+0x1c/0x3c [mv643xx_eth])
[   11.437319] [<bf04001c>] (mv643xx_eth_init_module+0x1c/0x3c [mv643xx_eth]) from [<c0008574>] (do_one_initcall+0x98/0x170)
[   11.448332] [<c0008574>] (do_one_initcall+0x98/0x170) from [<c005a288>] (sys_init_module+0x1710/0x18b4)
[   11.457777] [<c005a288>] (sys_init_module+0x1710/0x18b4) from [<c000dae0>] (ret_fast_syscall+0x0/0x2c)
[   11.467129] Code: e2800008 eb48d099 e5840400 e5951004 (e5903000) 
[   11.473269] ---[ end trace b1a3bb64d0c5b5ce ]---

[matth45]

I noticed with the 3.6.0 rc6 kernel that I get this message earlier in the boot process. I hadn't noticed it before, since it scrolled by quickly. Have you gotten anything like this? I wonder if its related (I didn't get anything like this with kernel-kirkwood-3.4.2-3.fc17.armv5tel).

[   11.172389] mv_xor mv_xor.0: DMA-API: device driver frees DMA memory with wrong function [device address=0x000000001f1ee040] [size=2000 bytes] [ma]
[   11.172397] Modules linked in:
[   11.172434] [<c000ee60>] (unwind_backtrace+0x0/0x124) from [<c0017d90>] (warn_slowpath_common+0x54/0x6c)
[   11.172456] [<c0017d90>] (warn_slowpath_common+0x54/0x6c) from [<c0017e40>] (warn_slowpath_fmt+0x34/0x44)
[   11.172476] [<c0017e40>] (warn_slowpath_fmt+0x34/0x44) from [<c026ae98>] (check_unmap+0x37c/0x754)
[   11.172494] [<c026ae98>] (check_unmap+0x37c/0x754) from [<c026b33c>] (debug_dma_unmap_page+0x64/0x70)
[   11.172515] [<c026b33c>] (debug_dma_unmap_page+0x64/0x70) from [<c02931b0>] (mv_xor_slot_cleanup+0x214/0x3b4)
[   11.172537] [<c02931b0>] (mv_xor_slot_cleanup+0x214/0x3b4) from [<c0021d48>] (tasklet_action+0x88/0xd8)
[   11.172555] [<c0021d48>] (tasklet_action+0x88/0xd8) from [<c0021090>] (__do_softirq+0x16c/0x394)
[   11.172573] [<c0021090>] (__do_softirq+0x16c/0x394) from [<c0021778>] (irq_exit+0x5c/0xc0)
[   11.172591] [<c0021778>] (irq_exit+0x5c/0xc0) from [<c0009ba8>] (handle_IRQ+0x6c/0x8c)
[   11.172610] [<c0009ba8>] (handle_IRQ+0x6c/0x8c) from [<c04df458>] (__irq_svc+0x38/0x80)
[   11.172629] [<c04df458>] (__irq_svc+0x38/0x80) from [<c04df078>] (_raw_spin_unlock_irqrestore+0x4c/0x64)
[   11.172650] [<c04df078>] (_raw_spin_unlock_irqrestore+0x4c/0x64) from [<c02606cc>] (__debug_object_init+0x30c/0x348)
[   11.172671] [<c02606cc>] (__debug_object_init+0x30c/0x348) from [<c00280d0>] (init_timer_on_stack_key+0x20/0x38)
[   11.172694] [<c00280d0>] (init_timer_on_stack_key+0x20/0x38) from [<c04db0b0>] (schedule_timeout+0x94/0x2f4)
[   11.172714] [<c04db0b0>] (schedule_timeout+0x94/0x2f4) from [<c00295a0>] (msleep+0x1c/0x28)
[   11.172732] [<c00295a0>] (msleep+0x1c/0x28) from [<c04d1ea8>] (mv_xor_probe+0x4f0/0xb48)
[   11.172753] [<c04d1ea8>] (mv_xor_probe+0x4f0/0xb48) from [<c02c6230>] (platform_drv_probe+0x1c/0x20)
[   11.172773] [<c02c6230>] (platform_drv_probe+0x1c/0x20) from [<c02c4fd8>] (driver_probe_device+0x12c/0x348)
[   11.172791] [<c02c4fd8>] (driver_probe_device+0x12c/0x348) from [<c02c526c>] (__driver_attach+0x78/0x9c)
[   11.172809] [<c02c526c>] (__driver_attach+0x78/0x9c) from [<c02c3390>] (bus_for_each_dev+0x50/0x88)
[   11.172827] [<c02c3390>] (bus_for_each_dev+0x50/0x88) from [<c02c454c>] (bus_add_driver+0x108/0x268)
[   11.172845] [<c02c454c>] (bus_add_driver+0x108/0x268) from [<c02c5568>] (driver_register+0xa4/0x134)
[   11.172867] [<c02c5568>] (driver_register+0xa4/0x134) from [<c06cfc7c>] (mv_xor_init+0x1c/0x3c)
[   11.172885] [<c06cfc7c>] (mv_xor_init+0x1c/0x3c) from [<c000859c>] (do_one_initcall+0x98/0x174)
[   11.172907] [<c000859c>] (do_one_initcall+0x98/0x174) from [<c06b2974>] (kernel_init+0x138/0x1f8)
[   11.172926] [<c06b2974>] (kernel_init+0x138/0x1f8) from [<c0009c6c>] (kernel_thread_exit+0x0/0x8)
[   11.173153] ---[ end trace cacf76ca10360dbf ]---
[   11.173163] Mapped at:
[   11.173171]  [<c026b94c>] debug_dma_map_page+0x40/0x110
[   11.173190]  [<c04d1dbc>] mv_xor_probe+0x404/0xb48
[   11.173207]  [<c02c6230>] platform_drv_probe+0x1c/0x20
[   11.173223]  [<c02c4fd8>] driver_probe_device+0x12c/0x348
[   11.173238]  [<c02c526c>] __driver_attach+0x78/0x9c
[   11.205383] mv_xor mv_xor.0: Marvell XOR: ( xor cpy )
[   11.245376] mv_xor mv_xor.1: Marvell XOR: ( xor fill cpy )
[   11.285373] mv_xor mv_xor.2: Marvell XOR: ( xor cpy )
[   11.325377] mv_xor mv_xor.3: Marvell XOR: ( xor fill cpy )

---------- Post added at 11:47 PM ---------- Previous post was at 11:17 PM ----------

I was thinking about what you said about u-boot. I shouldn't be so quick to rule it out (I'm really kind of a novice at this). I do wonder how fedora chose the memory addresses to load the kernel image and initramfs. These addresses are interpreted by u-boot, so perhaps some versions behave differently? If part of one of these images were overwritten because they were incorrectly spaced in memory by u-boot, I would expect it to create a problem.

At any rate, I have a u-boot from Marvell, version 3.4.27. Does your u-boot boot message look anything like this?

The message below says that I have 512MB of memory exactly, split between two 256MB modules, with the address of each module starting at 0x00000000 and 0x10000000 (0MB and 256MB). Then 0x06400000 is loading the kernel image starting at 100MB, and the initial RAM filesystem at 116MB (plenty of room between 100MB and 116MB, since the uImage-kirkwood is less than 4MB and is not compressed). It also seems like the first 8MB are being reserved for u-boot. So what happens to the memory between 8MB and 100MB?

When I boot the system /proc/meminfo says there are 491MB of memory total (~21MB less than the 512MB on the system). So are the u-boot images ever freed?

         __  __                      _ _
        |  \/  | __ _ _ ____   _____| | |
        | |\/| |/ _` | '__\ \ / / _ \ | |
        | |  | | (_| | |   \ V /  __/ | |
        |_|  |_|\__,_|_|    \_/ \___|_|_|
 _   _     ____              _
| | | |   | __ )  ___   ___ | |_ 
| | | |___|  _ \ / _ \ / _ \| __| 
| |_| |___| |_) | (_) | (_) | |_ 
 \___/    |____/ \___/ \___/ \__| 
 ** MARVELL BOARD: SHEEVA PLUG LE 

U-Boot 1.1.4 (Dec 27 2009 - 22:03:21) Marvell version: 3.4.27

U-Boot code: 00600000 -> 0067FFF0  BSS: -> 006CFEE0

Soc: 88F6281 A0 (DDR2)
CPU running @ 1200Mhz L2 running @ 400Mhz
SysClock = 400Mhz , TClock = 200Mhz 

DRAM CAS Latency = 5 tRP = 5 tRAS = 18 tRCD=6
DRAM CS[0] base 0x00000000   size 256MB 
DRAM CS[1] base 0x10000000   size 256MB 
DRAM Total size 512MB  16bit width
Addresses 8M - 0M are saved for the U-Boot usage.
Mem malloc Initialization (8M - 7M): Done
NAND:512 MB
Flash:  0 kB

CPU : Marvell Feroceon (Rev 1)

Streaming disabled 
Write allocate disabled


USB 0: host mode
PEX 0: interface detected no Link.
Net:   egiga0 [PRIME]
Hit any key to stop autoboot:  0

[efloehr]

Here is my U-Boot initialization:

U-Boot 2011.06 (Oct 15 2011 - 02:02:08)
Marvell-DreamPlug

SoC:   Kirkwood 88F6281_A0
DRAM:  512 MiB
SF: Detected MX25L1606 with page size 256, total 1 MiB
In:    serial
Out:   serial
Err:   serial
Net:   egiga0, egiga1
88E1121 Initialized on egiga0
88E1121 Initialized on egiga1
Hit any key to stop autoboot:  0

The pre-installed OS is Debian 6.0.2 with a kernel apparently compiled by Global Scale:

root@dreamplug-debian:~/firmware# uname -a
Linux dreamplug-debian 2.6.39.4 #62 PREEMPT Sat Mar 31 16:14:14 EDT 2012 armv5tel GNU/Linux

The ethernet driver loaded is:

[   16.423944] mv643xx_eth: MV-643xx 10/100/1000 ethernet driver version 1.4

Global Scale is definitely compiling something proprietary into the kernel:

[   65.200647] mlan: module license 'Marvell Proprietary' taints kernel.
[   65.207126] Disabling lock debugging due to kernel taint

Anyway, just some nuggets...still no luck getting Fedora to work.

Eric

[efloehr]

All,

After posting on the Fedora ARM mailing list, with their help, it was determined that the boot issue with my Dreamplug was Fedora trying to load the orion_nand kernel module. The Dreamplug does not have NAND, so it was hanging.

So I guess the next step is to see if the nand module can smartly not load, or at least not hand when nand isn't present.

Best,
Eric