Message-ID: <20140520195041.GA28913@roeck-us.net>
Date:	Tue, 20 May 2014 12:50:41 -0700
From:	Guenter Roeck <linux@...ck-us.net>
To:	Francesco Ruggeri <fruggeri@...sta.com>
Cc:	linux-pci@...r.kernel.org, linux-kernel@...r.kernel.org,
	Francesco Ruggeri <fruggeri@...stanetworks.com>
Subject: Re: pci: kernel crash in bus_find_device

On Tue, May 20, 2014 at 12:17:57PM -0700, Francesco Ruggeri wrote:
> I posted this about a week ago but I did not get any replies.
> Re-trying.
> 
> While traversing devices on pci_bus_type I ran into the crash below.
> The immediate cause of the crash is that bus_find_device is trying to resume
> a scan starting from a device that has been unregistered (and whose knode_bus
> has already been klist_del' ed).
> The main issue seems to be that when resuming a scan the caller should
> be holding a
> reference to the klist_node, but instead it relies on holding a
> reference to the device.
> I played with a couple of narrow fixes, but a clean solution would
> affect quite a bit of code.
> 
> Has anybody run into this before?
> 

Hi Francesco,

I may be missing something, but I don't find a pci_scan symbol in the 3.4
kernel. Also, the process name suggests that you may possibly trigger pci
rescans from user space. Both suggest that you may possibly run third party
code in your kernel.

Either case, I ran into similar problems myself with pci rescans triggered
from user space. The 3.4 kernel has no synchronization for rescans triggered
from user space with those triggered from the kernel. In a nutshell, when
triggering rescans and removals from user space you must ensure that only
one such rescan/removal is active at any given time. Under no circumstances
trigger rescans from user space if a rescan can also be triggered from the
kernel. Obviously that also applies if rescans can be triggered multiple times
in parallel by some third party kernel module. Maybe that explains your
problem ?

The problem has been addressed recently with commit 9d16947 (PCI: Add
global pci_lock_rescan_remove) and several subsequent patches.

Guenter

> Thanks,
> Francesco Ruggeri
> 
> 
> ------------[ cut here ]------------
> WARNING: at /bld/EosKernel/Artools-rpmbuild/linux-3.4/include/linux/kref.h:41
> klist_iter_init_node+0x30/0x38()
> Modules linked in: pci_scan(O) sch_prio sand_dma(PO) arista_bde(PO)
> macvlan ip6table_mangle iptable_mangle msr nf_conntrack_ipv6
> nf_defrag_ipv6 ip6t_REJECT nf_conntrack_ipv4 nf_defrag_ipv4 xt_LOG
> xt_limit ipt_REJECT xt_hl xt_state xt_multiport xt_tcpudp kbfd(O)
> 8021q garp stp llc tun scd_em_driver(O) nf_conntrack_tftp iptable_raw
> iptable_filter ip_tables xt_NOTRACK nf_conntrack xt_mark ip6table_raw
> ip6table_filter ip6_tables x_tables scd(O) k8temp amd64_edac_mod hwmon
> kvm_amd kvm
> Pid: 6861, comm: pci_scan_0 Tainted: P           O
> 3.4.43.Ar-1797671.flbocafruggeri #1
> Call Trace:
>  [<ffffffff81029dc4>] warn_slowpath_common+0x80/0x98
>  [<ffffffff811b57f1>] ? pci_do_find_bus+0x49/0x49
>  [<ffffffff81029df1>] warn_slowpath_null+0x15/0x17
>  [<ffffffff813a43ce>] klist_iter_init_node+0x30/0x38
>  [<ffffffff8120e57e>] bus_find_device+0x48/0x90
>  [<ffffffff811b5908>] pci_get_dev_by_id+0x5e/0x81
>  [<ffffffff811b5a6a>] pci_get_subsys+0x5c/0x7f
>  [<ffffffff811b5a9e>] pci_get_device+0x11/0x13
>  [<ffffffffa00b2087>] pci_scan+0x39/0x8a [pci_scan]
>  [<ffffffffa00b204e>] ? init_module+0x3c/0x3c [pci_scan]
>  [<ffffffff81040e6e>] kthread+0x84/0x8c
>  [<ffffffff813c8b14>] kernel_thread_helper+0x4/0x10
>  [<ffffffff81040dea>] ? __init_kthread_worker+0x37/0x37
>  [<ffffffff813c8b10>] ? gs_change+0xb/0xb
> ---[ end trace 79cea1ec476672fe ]---
> ------------[ cut here ]------------
> WARNING: at /bld/EosKernel/Artools-rpmbuild/linux-3.4/lib/klist.c:189
> klist_release+0x2b/0xeb()
> Modules linked in: pci_scan(O) sch_prio sand_dma(PO) arista_bde(PO)
> macvlan ip6table_mangle iptable_mangle msr nf_conntrack_ipv6
> nf_defrag_ipv6 ip6t_REJECT nf_conntrack_ipv4 nf_defrag_ipv4 xt_LOG
> xt_limit ipt_REJECT xt_hl xt_state xt_multiport xt_tcpudp kbfd(O)
> 8021q garp stp llc tun scd_em_driver(O) nf_conntrack_tftp iptable_raw
> iptable_filter ip_tables xt_NOTRACK nf_conntrack xt_mark ip6table_raw
> ip6table_filter ip6_tables x_tables scd(O) k8temp amd64_edac_mod hwmon
> kvm_amd kvm
> Pid: 6861, comm: pci_scan_0 Tainted: P        W  O
> 3.4.43.Ar-1797671.flbocafruggeri #1
> Call Trace:
>  [<ffffffff81029dc4>] warn_slowpath_common+0x80/0x98
>  [<ffffffff8120de13>] ? bus_get_device_klist+0x10/0x10
>  [<ffffffff81029df1>] warn_slowpath_null+0x15/0x17
>  [<ffffffff813a440e>] klist_release+0x2b/0xeb
>  [<ffffffff813a44ec>] klist_dec_and_del+0x1e/0x25
>  [<ffffffff813a4528>] klist_next+0x35/0xc9
>  [<ffffffff811b57f1>] ? pci_do_find_bus+0x49/0x49
>  [<ffffffff8120deb3>] next_device+0x9/0x19
>  [<ffffffff8120e5a2>] bus_find_device+0x6c/0x90
>  [<ffffffff811b5908>] pci_get_dev_by_id+0x5e/0x81
>  [<ffffffff811b5a6a>] pci_get_subsys+0x5c/0x7f
>  [<ffffffff811b5a9e>] pci_get_device+0x11/0x13
>  [<ffffffffa00b2087>] pci_scan+0x39/0x8a [pci_scan]
>  [<ffffffffa00b204e>] ? init_module+0x3c/0x3c [pci_scan]
>  [<ffffffff81040e6e>] kthread+0x84/0x8c
>  [<ffffffff813c8b14>] kernel_thread_helper+0x4/0x10
>  [<ffffffff81040dea>] ? __init_kthread_worker+0x37/0x37
>  [<ffffffff813c8b10>] ? gs_change+0xb/0xb
> ---[ end trace 79cea1ec476672ff ]---
> general protection fault: 0000 [#1] PREEMPT SMP
> CPU 1
> Modules linked in: pci_scan(O) sch_prio sand_dma(PO) arista_bde(PO)
> macvlan ip6table_mangle iptable_mangle msr nf_conntrack_ipv6
> nf_defrag_ipv6 ip6t_REJECT nf_conntrack_ipv4 nf_defrag_ipv4 xt_LOG
> xt_limit ipt_REJECT xt_hl xt_state xt_multiport xt_tcpudp kbfd(O)
> 8021q garp stp llc tun scd_em_driver(O) nf_conntrack_tftp iptable_raw
> iptable_filter ip_tables xt_NOTRACK nf_conntrack xt_mark ip6table_raw
> ip6table_filter ip6_tables x_tables scd(O) k8temp amd64_edac_mod hwmon
> kvm_amd kvm
> 
> Pid: 6861, comm: pci_scan_0 Tainted: P        W  O
> 3.4.43.Ar-1797671.flbocafruggeri #1
> RIP: 0010:[<ffffffff813a442c>]  [<ffffffff813a442c>] klist_release+0x49/0xeb
> RSP: 0018:ffff88001c55bd50  EFLAGS: 00010293
> RAX: dead000000200200 RBX: ffff880030949e78 RCX: ffff880000000010
> RDX: dead000000100100 RSI: 0000000000000000 RDI: dead000000200200
> RBP: ffff88001c55bd70 R08: dead000000100100 R09: 000000000000000a
> R10: 0000000000000000 R11: ffffffff81619920 R12: ffff880030949e90
> R13: ffff880030949e78 R14: ffffffff8120de13 R15: ffff880027e717e0
> FS:  0000000000000000(0000) GS:ffff88013fb00000(0000) knlGS:00000000f73bc6d0
> CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
> CR2: 0000000009012644 CR3: 0000000069f9e000 CR4: 00000000000007e0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
> Process pci_scan_0 (pid: 6861, threadinfo ffff88001c55a000, task
> ffff880032ffd340)
> Stack:
>  ffff880030949e78 ffff88001c55bde0 dead000000100100 ffff880030949e78
>  ffff88001c55bd80 ffffffff813a44ec ffff88001c55bdc0 ffffffff813a4528
>  ffff88001c55bde0 ffff880027e717e0 ffffffff811b57f1 ffff88001c55bde0
> Call Trace:
>  [<ffffffff813a44ec>] klist_dec_and_del+0x1e/0x25
>  [<ffffffff813a4528>] klist_next+0x35/0xc9
>  [<ffffffff811b57f1>] ? pci_do_find_bus+0x49/0x49
>  [<ffffffff8120deb3>] next_device+0x9/0x19
>  [<ffffffff8120e5a2>] bus_find_device+0x6c/0x90
>  [<ffffffff811b5908>] pci_get_dev_by_id+0x5e/0x81
>  [<ffffffff811b5a6a>] pci_get_subsys+0x5c/0x7f
>  [<ffffffff811b5a9e>] pci_get_device+0x11/0x13
>  [<ffffffffa00b2087>] pci_scan+0x39/0x8a [pci_scan]
>  [<ffffffffa00b204e>] ? init_module+0x3c/0x3c [pci_scan]
>  [<ffffffff81040e6e>] kthread+0x84/0x8c
>  [<ffffffff813c8b14>] kernel_thread_helper+0x4/0x10
>  [<ffffffff81040dea>] ? __init_kthread_worker+0x37/0x37
>  [<ffffffff813c8b10>] ? gs_change+0xb/0xb
> Code: 00 48 c7 c7 a1 01 51 81 e8 ce 59 c8 ff 49 8b 54 24 f0 49 8b 44
> 24 f8 49 b8 00 01 10 00 00 00 ad de 48 bf 00 02 20 00 00 00 ad de <48>
> 89 42 08 48 89 10 49 89 7c 24 f8 4d 89 44 24 f0 48 c7 c7 30
> RIP  [<ffffffff813a442c>] klist_release+0x49/0xeb
>  RSP <ffff88001c55bd50>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@...r.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
> 
> 
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists