Re: mmotm 2009-09-09-22-56 uploaded

[Suresh Jayaram]

On Wed, Sep 16, 2009 at 4:09 AM, David Rientjes <rientjes@xxxxxxxxxx> wrote:
> On Tue, 15 Sep 2009, Suresh Jayaram wrote:
>
>> >  [ I notice Thomas has since provided patches for this problem, so I'll
>> >   check if it solves this issue on T60p. ]
>> >
>
> Thomas' patches in timers/core do fix this issue, so no revert will be
> necessary in linux-next when the next -mm is released.
>
>> Sep 15 15:29:36 jay-t60p kernel: BUG: unable to handle kernel NULL
>> pointer dereference at (null)
>> Sep 15 15:29:36 jay-t60p kernel: IP: [<ffffffff81178954>]
>> bitmap_scnprintf+0x5a/0xae
>> Sep 15 15:29:36 jay-t60p kernel: PGD 0
>> Sep 15 15:29:36 jay-t60p kernel: Oops: 0000 [#1] SMP
>> Sep 15 15:29:36 jay-t60p kernel: last sysfs file:
>> /sys/devices/pci0000:00/0000:00:01.0/local_cpus
>> Sep 15 15:29:36 jay-t60p kernel: CPU 1
>> Sep 15 15:29:36 jay-t60p kernel: Modules linked in: bnep sco nfsd
>> exportfs rfcomm l2cap crc16 snd_pcm_oss snd_mixer_oss snd_seq
>> binfmt_misc snd_seq_device nfs lockd nfs_acl auth_rpcgss sunrpc ipv6
>> cpufreq_conservative cpufreq_userspace cpufreq_powersave acpi_cpufreq
>> ipt_MASQUERADE iptable_nat nf_nat nf_conntrack_ipv4 nf_conntrack
>> nf_defrag_ipv4 ip_tables x_tables bridge stp tun fuse loop dm_mod
>> snd_hda_codec_analog snd_hda_intel snd_hda_codec snd_hwdep rtc_cmos
>> pcmcia btusb snd_pcm snd_timer snd ppdev thinkpad_acpi rtc_core
>> i2c_i801 yenta_socket iTCO_wdt nsc_ircc usbhid ide_cd_mod video
>> rsrc_nonstatic pl2303 bluetooth hid soundcore parport_pc led_class
>> output rtc_lib ac battery button intel_agp i2c_core pcmcia_core pcspkr
>> usbserial iTCO_vendor_support irda e1000e cdrom snd_page_alloc parport
>> sg rfkill joydev uinput crc_ccitt uhci_hcd ehci_hcd sd_mod crc_t10dif
>> usbcore edd ext3 mbcache jbd fan ide_pci_generic ata_generic ata_piix
>> pata_acpi ahci libata scsi_mod piix ide_core thermal processor
>> thermal_sys hwmon
>> Sep 15 15:29:36 jay-t60p kernel: Pid: 2478, comm: irqbalance Not
>> tainted 2.6.31-rc9-mm1-0.1-default #7 20088JQ
>> Sep 15 15:29:36 jay-t60p kernel: RIP: 0010:[<ffffffff81178954>]
>> [<ffffffff81178954>] bitmap_scnprintf+0x5a/0xae
>> Sep 15 15:29:36 jay-t60p kernel: RSP: 0018:ffff880037a81e38  EFLAGS: 00010246
>> Sep 15 15:29:36 jay-t60p kernel: RAX: 0000000000000000 RBX:
>> 0000000000000000 RCX: 0000000000000000
>> Sep 15 15:29:36 jay-t60p kernel: RDX: 0000000000000002 RSI:
>> 0000000000000ffe RDI: 0000000000000000
>> Sep 15 15:29:36 jay-t60p kernel: RBP: ffff880037a81e68 R08:
>> 0000000000000005 R09: 0000000000000001
>> Sep 15 15:29:36 jay-t60p kernel: R10: ffffffff813c1cd3 R11:
>> 0000000000000001 R12: 0000000000000000
>> Sep 15 15:29:36 jay-t60p kernel: R13: 0000000000000000 R14:
>> 0000000000000ffe R15: ffff88007bbe3000
>> Sep 15 15:29:36 jay-t60p kernel: FS:  00007f605bee46f0(0000)
>> GS:ffff880005f00000(0000) knlGS:0000000000000000
>> Sep 15 15:29:36 jay-t60p kernel: CS:  0010 DS: 0000 ES: 0000 CR0:
>> 000000008005003b
>> Sep 15 15:29:36 jay-t60p kernel: CR2: 0000000000000000 CR3:
>> 000000007acf8000 CR4: 00000000000006e0
>> Sep 15 15:29:36 jay-t60p kernel: DR0: 0000000000000000 DR1:
>> 0000000000000000 DR2: 0000000000000000
>> Sep 15 15:29:36 jay-t60p kernel: DR3: 0000000000000000 DR6:
>> 00000000ffff0ff0 DR7: 0000000000000400
>> Sep 15 15:29:36 jay-t60p kernel: Process irqbalance (pid: 2478,
>> threadinfo ffff880037a80000, task ffff8800376d8280)
>> Sep 15 15:29:37 jay-t60p rpc.statd[2588]: statd running as root. chown
>> /var/lib/nfs/sm to choose different user
>> Sep 15 15:29:37 jay-t60p sm-notify[2596]: Already notifying clients; Exiting!
>> Sep 15 15:29:36 jay-t60p modem-manager: Loaded plugin MBM
>> Sep 15 15:29:36 jay-t60p kernel: Stack:
>> Sep 15 15:29:37 jay-t60p kernel:  ffffffff8152b7d0 ffff88007bbe3000
>> ffffffff8154a6a8 ffff88007c7aa2d8
>> Sep 15 15:29:37 jay-t60p kernel: <0> ffffffff81556620 ffff880037a81f48
>> ffff880037a81e88 ffffffff81186937
>> Sep 15 15:29:37 jay-t60p kernel: <0> ffff88007c7aa2d8 fffffffffffffffb
>> ffff880037a81ea8 ffffffff811fbff6
>> Sep 15 15:29:37 jay-t60p kernel: Call Trace:
>> Sep 15 15:29:37 jay-t60p kernel:  [<ffffffff81186937>] local_cpus_show+0x36/0x49
>> Sep 15 15:29:37 jay-t60p kernel:  [<ffffffff811fbff6>] dev_attr_show+0x20/0x44
>> Sep 15 15:29:37 jay-t60p kernel:  [<ffffffff81121741>]
>> sysfs_read_file+0xb2/0x131
>> Sep 15 15:29:37 jay-t60p kernel:  [<ffffffff810ce07c>] vfs_read+0xb1/0x187
>> Sep 15 15:29:37 jay-t60p kernel:  [<ffffffff810ce216>] sys_read+0x47/0x70
>> Sep 15 15:29:37 jay-t60p modem-manager: Loaded plugin Sierra
>> Sep 15 15:29:37 jay-t60p kernel:  [<ffffffff8100b9eb>]
>> system_call_fastpath+0x16/0x1b
>
> Looks like a bootstrap issue since is appears as though `local_cpus' is
> being read by some initscript on your T60p and mine doesn't suffer from
> the same problem (it isn't read at all on mine).
>
> CONFIG_DEBUG_PER_CPU_MAPS should be able to diagnose it when you get back
> from traveling.
>

I enabled the CONFIG_DEBUG_PER_CPU_MAPS and rebuilt the kernel, but
now I don't see the oops anymore.
But what I see in dmesg output is:

Sep 16 09:22:57 jay-t60p kernel: cpumask_of_node(-1): no node_to_cpumask_map!

followed by a call trace for irqbalance (attached).

Thanks,
Suresh Jayaraman

Attachment: dmesg-irqbalance
Description: Binary data