lkml.org 
[lkml]   [2009]   [Jul]   [2]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
Subjectlockdep max numbers (was Re: Crashes during boot on 2.6.30 / 2.6.31-rc, random programs)
On Wed, Jul 01, 2009 at 09:30:23PM +0100, Joao Correia wrote:
>Hello again
>
>Looks like this just opened the lid on some other limits. I just hit
>another thing, which hadn't shown up before i made the changes, but
>looks like its just another limit thats too low. This one is harder to
>reproduce tho.
>
>Jun 30 21:35:03 hightech kernel: BUG: MAX_LOCKDEP_CHAINS too low!
>Jun 30 21:35:03 hightech kernel: turning off the locking correctness validator.
>Jun 30 21:35:03 hightech kernel: Pid: 9379, comm: qemu-kvm Not tainted


Similar problem...

Peter? Would like to increase these numbers?


>2.6.30-wl #3
>Jun 30 21:35:03 hightech kernel: Call Trace:
>Jun 30 21:35:03 hightech kernel: [<c07da76f>] ? printk+0x22/0x3b
>Jun 30 21:35:03 hightech kernel: [<c04638f5>] __lock_acquire+0x583/0xb05
>Jun 30 21:35:03 hightech kernel: [<c046a107>] ? __module_text_address+0x20/0x6d
>Jun 30 21:35:03 hightech kernel: [<c0463f2e>] lock_acquire+0xb7/0xeb
>Jun 30 21:35:03 hightech kernel: [<c05cf959>] ? get_hash_bucket+0x3a/0x55
>Jun 30 21:35:03 hightech kernel: [<c05cf959>] ? get_hash_bucket+0x3a/0x55
>Jun 30 21:35:03 hightech kernel: [<c07dd75f>] _spin_lock_irqsave+0x45/0x89
>Jun 30 21:35:03 hightech kernel: [<c05cf959>] ? get_hash_bucket+0x3a/0x55
>Jun 30 21:35:03 hightech kernel: [<c05cf959>] get_hash_bucket+0x3a/0x55
>Jun 30 21:35:03 hightech kernel: [<c05d09e8>] add_dma_entry+0x1f/0x4f
>Jun 30 21:35:03 hightech kernel: [<c05d0c66>] debug_dma_map_sg+0xe1/0x147
>Jun 30 21:35:03 hightech kernel: [<c0695a19>] ata_qc_issue+0x1c0/0x254
>Jun 30 21:35:03 hightech kernel: [<c069bf44>] __ata_scsi_queuecmd+0x16a/0x1c8
>Jun 30 21:35:03 hightech kernel: [<c067a37e>] ? scsi_done+0x0/0x32
>Jun 30 21:35:03 hightech kernel: [<c069b9b1>] ? ata_scsi_rw_xlat+0x0/0x207
>Jun 30 21:35:03 hightech kernel: [<c069c05b>] ata_scsi_queuecmd+0x55/0x95
>Jun 30 21:35:03 hightech kernel: [<c067a37e>] ? scsi_done+0x0/0x32
>Jun 30 21:35:03 hightech kernel: [<c067a627>] scsi_dispatch_cmd+0x187/0x207
>Jun 30 21:35:03 hightech kernel: [<c068044c>] scsi_request_fn+0x33d/0x480
>Jun 30 21:35:03 hightech kernel: [<c0445a1e>] ? del_timer+0x54/0x6e
>Jun 30 21:35:03 hightech kernel: [<c05aefbe>] __generic_unplug_device+0x3e/0x53
>Jun 30 21:35:03 hightech kernel: [<c05af00a>] blk_start_queueing+0x37/0x4a
>Jun 30 21:35:03 hightech kernel: [<c05bcc7e>] cfq_insert_request+0x46f/0x498
>Jun 30 21:35:03 hightech kernel: [<c05ad43b>] ? __rcu_read_unlock+0x0/0x3d
>Jun 30 21:35:03 hightech kernel: [<c05ac60d>] elv_insert+0x12f/0x1d0
>Jun 30 21:35:03 hightech kernel: [<c05ac763>] __elv_add_request+0xb5/0xce
>Jun 30 21:35:03 hightech kernel: [<c05ad465>] ? __rcu_read_unlock+0x2a/0x3d
>Jun 30 21:35:03 hightech kernel: [<c05ad533>] ? drive_stat_acct+0xbb/0xd3
>Jun 30 21:35:03 hightech kernel: [<c05af6b3>] __make_request+0x33b/0x3c1
>Jun 30 21:35:03 hightech kernel: [<c04cd85c>] ? check_valid_pointer+0x2c/0x6c
>Jun 30 21:35:03 hightech kernel: [<c046249e>] ? mark_lock+0x29/0x1f3
>Jun 30 21:35:03 hightech kernel: [<c04cf83a>] ? kmem_cache_alloc+0xc2/0x150
>Jun 30 21:35:03 hightech kernel: [<c05ade19>] generic_make_request+0x2fd/0x351
>Jun 30 21:35:03 hightech kernel: [<c05ade19>] ? generic_make_request+0x2fd/0x351
>Jun 30 21:35:03 hightech kernel: [<c04a8b4e>] ? mempool_alloc+0x5b/0x105
>Jun 30 21:35:03 hightech kernel: [<c04626ba>] ? mark_held_locks+0x52/0x7c
>Jun 30 21:35:03 hightech kernel: [<c05adf2d>] submit_bio+0xc0/0xd9
>Jun 30 21:35:03 hightech kernel: [<c04fd28d>] ? bio_alloc_bioset+0x35/0xd5
>Jun 30 21:35:03 hightech kernel: [<c04f845c>] submit_bh+0xea/0x11b
>Jun 30 21:35:03 hightech kernel: [<c04fae00>]
>__block_write_full_page+0x218/0x2f0
>Jun 30 21:35:03 hightech kernel: [<c054427d>] ?
>ext4_normal_get_block_write+0x0/0x76
>Jun 30 21:35:03 hightech kernel: [<c04fb5ea>] ? end_buffer_async_write+0x0/0x140
>Jun 30 21:35:03 hightech kernel: [<c04fb5ea>] ? end_buffer_async_write+0x0/0x140
>Jun 30 21:35:03 hightech kernel: [<c04faf4c>]
>block_write_full_page_endio+0x74/0xda
>Jun 30 21:35:03 hightech kernel: [<c04fb5ea>] ? end_buffer_async_write+0x0/0x140
>Jun 30 21:35:03 hightech kernel: [<c054427d>] ?
>ext4_normal_get_block_write+0x0/0x76
>Jun 30 21:35:03 hightech kernel: [<c04fafd4>] block_write_full_page+0x22/0x38
>Jun 30 21:35:03 hightech kernel: [<c04fb5ea>] ? end_buffer_async_write+0x0/0x140
>Jun 30 21:35:03 hightech kernel: [<c0543390>] ext4_da_writepage+0x1aa/0x1da
>Jun 30 21:35:03 hightech kernel: [<c054315b>] mpage_da_submit_io+0x9b/0xf0
>Jun 30 21:35:03 hightech kernel: [<c0545f40>] ext4_da_writepages+0x2c2/0x445
>Jun 30 21:35:03 hightech kernel: [<c04ad194>] do_writepages+0x32/0x58
>Jun 30 21:35:03 hightech kernel: [<c04a739b>]
>__filemap_fdatawrite_range+0x73/0x8c
>Jun 30 21:35:03 hightech kernel: [<c04a7c81>] filemap_fdatawrite_range+0x2b/0x44
>Jun 30 21:35:03 hightech kernel: [<c04a7e9e>] sync_page_range+0x7f/0xf5
>Jun 30 21:35:03 hightech kernel: [<c04629d3>] ? trace_hardirqs_on+0x19/0x2c
>Jun 30 21:35:03 hightech kernel: [<c04a7fd1>] generic_file_aio_write+0xbd/0xe1
>Jun 30 21:35:03 hightech kernel: [<c053da09>] ext4_file_write+0xc2/0x149
>Jun 30 21:35:03 hightech kernel: [<c04d7cf2>] do_sync_write+0xbe/0x10a
>Jun 30 21:35:03 hightech kernel: [<c046249e>] ? mark_lock+0x29/0x1f3
>Jun 30 21:35:03 hightech kernel: [<c0451586>] ?
>autoremove_wake_function+0x0/0x55
>Jun 30 21:35:03 hightech kernel: [<c04d920e>] ? __rcu_read_unlock+0x0/0x3d
>Jun 30 21:35:03 hightech kernel: [<c058377a>] ?
>security_file_permission+0x23/0x36
>Jun 30 21:35:03 hightech kernel: [<c04d7eed>] ? rw_verify_area+0xa5/0xd8
>Jun 30 21:35:03 hightech kernel: [<c04d7c34>] ? do_sync_write+0x0/0x10a
>Jun 30 21:35:03 hightech kernel: [<c04d837f>] vfs_write+0x9f/0x10f
>Jun 30 21:35:03 hightech kernel: [<c04d8459>] sys_pwrite64+0x6a/0x93
>Jun 30 21:35:03 hightech kernel: [<c04037f3>] sysenter_do_call+0x12/0x38
>
>
>Now the whole call trace is showing up (so the other limits were
>raised correctly).
>
>Im going to go through the source and raise this limit as well, but as
>i said above, this one is harder to trigger, so it may take a couple
>of days to see it again.
>
>The first problem was indeed just the tip of the iceberg. Perphaps a
>more comprehensive change of all the related limits is in order? (I do
>not know if such change has already happened or not, if so, please
>point me in the right direction.)
>
>Again,
>Thank you very much for your time,
>Joao Correia
>Centro de Informatica
>Universidade da Beira Interior
>Portugal
>
>On Tue, Jun 30, 2009 at 10:16 AM, Amerigo Wang<xiyou.wangcong@gmail.com> wrote:
>> On Mon, Jun 29, 2009 at 11:30:41AM +0100, Joao Correia wrote:
>>>Hello
>>>
>>>The system seemed to happily ignore all the sysctl.conf changes and
>>>all echo VALUE >  /proc/sys/kernel/max_lock_depth
>>
>>
>> Ouch, mistake! max_lock_depth is for rt_mutex... :)
>>
>>>
>>>So i dug a little on the source and changed
>>>
>>>include/linux/sched.h
>>>
>>># define MAX_LOCK_DEPTH 48UL
>>>
>>>to
>>>
>>># define MAX_LOCK_DEPTH 96UL
>>>
>>>and im getting no more errors. Of course, now its probably radioactive
>>>and about to blow up, but at least it's not complaining anymore :).
>>
>> Yeah, this is right.
>>
>> Let's Cc: Peter to see if he would like to change this number...
>>
>> Peter?
>>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2009-07-02 11:31    [W:0.102 / U:0.196 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site