Re: [PATCH 0/3] PCI: fix the object lifetime issue of parallel device removal on different pci hierarchy

[Gu Zheng <guz.fnst@cn.fujitsu.com>]

On 04/25/2013 07:48 AM, Yinghai Lu wrote:

> On Wed, Apr 24, 2013 at 10:11 AM, Yinghai Lu <yinghai@kernel.org> wrote:
>> On Wed, Apr 24, 2013 at 2:36 AM, Gu Zheng <guz.fnst@cn.fujitsu.com> wrote:
>>> On 04/19/2013 03:43 AM, Yinghai Lu wrote:
>> [...]
>>>> I can not find where that reference count get reduced!
>>>> in the my test patches that I sent, it does have have that in remove path.
>>>>
>>>> What is reason for you to remove that line?
>>>>
>>>> After put the line back, your test still can pass?
>>>
>>> Hi Yinghai,
>>>         The remove test script still can pass if we reduce the reference of pci_bus when destroying
>>> the pci_dev, so does your patch, but a list_del corruption warning occurs in the bottom half routine:
>>
>> That is not clear.
>>
>> your patchset does not have list_del corruption warning?

Hi Yinghai,
	Sorry for my unclear description, my patchset has the same warning.

>>
>>> ------------[ cut here ]------------
>>> WARNING: at lib/list_debug.c:53 __list_del_entry+0x63/0xd0()
>>> Hardware name: PRIMEQUEST 1800E
>>> list_del corruption, ffff8807d1b6c000->next is LIST_POISON1 (dead000000100100)
>>> Modules linked in: shpchp ebtable_nat ebtables xt_CHECKSUM iptable_mangle bridge stp llc autofs4 sunrpc cpufreq_ondemand ipt_REJECT nf_conntrack_ipv4 nf_defrag_ipv4 iptable_filter ip_tables ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6 xt_state nf_conntrack ip6table_filter ip6_tables ipv6 vfat fat dm_mirror dm_region_hash dm_log dm_mod vhost_net macvtap macvlan tun uinput iTCO_wdt iTCO_vendor_support acpi_cpufreq freq_table mperf coretemp kvm_intel kvm crc32c_intel microcode pcspkr sg i2c_i801 lpc_ich mfd_core i7core_edac edac_core ioatdma e1000e igb dca i2c_algo_bit i2c_core ptp pps_core ext4(F) mbcache(F) jbd2(F) sd_mod(F) crc_t10dif(F) megaraid_sas(F) mptsas(F) mptscsih(F) mptbase(F) scsi_transport_sas(F)
>>> Pid: 6, comm: kworker/u:0 Tainted: GF       W    3.9.0-rc7-pci-remove-test+ #59
>>> Call Trace:
>>>  [<ffffffff81056d4f>] warn_slowpath_common+0x7f/0xc0
>>>  [<ffffffff81056e46>] warn_slowpath_fmt+0x46/0x50
>>>  [<ffffffff81280b13>] __list_del_entry+0x63/0xd0
>>>  [<ffffffff81280b91>] list_del+0x11/0x40
>>>  [<ffffffff81298331>] pci_destroy_dev+0x31/0xc0
>>>  [<ffffffff812985bb>] pci_remove_bus_device+0x5b/0x70
>>>  [<ffffffff812985ee>] pci_stop_and_remove_bus_device+0x1e/0x30
>>>  [<ffffffff8129fc89>] remove_callback+0x29/0x40
>>>  [<ffffffff811f3b84>] sysfs_schedule_callback_work+0x24/0x70
>>>  [<ffffffff81073d85>] process_one_work+0x185/0x3f0
>>>  [<ffffffff810763e9>] worker_thread+0x119/0x380
>>>  [<ffffffff810762d0>] ? manage_workers+0x180/0x180
>>>  [<ffffffff8107b6ae>] kthread+0xce/0xe0
>>>  [<ffffffff8107b5e0>] ? kthread_freezable_should_stop+0x70/0x70
>>>  [<ffffffff815724ac>] ret_from_fork+0x7c/0xb0
>>>  [<ffffffff8107b5e0>] ? kthread_freezable_should_stop+0x70/0x70
>>> ---[ end trace 9c05e382f933a515 ]---
> 
> Attached patch on top of current linus tree should fix the racing problem.
> 
> We don't need to bother to add get bus ref. etc.
> 
> Let us know your test result.

Thanks, I'll do the test soon, and send out the report later!

Best regards,
Gu

> 
> Yinghai