Re: BUG: unable to handle kernel paging request

[Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>]

On Mon, Apr 14, 2014 at 01:10:01PM +0000, Samuel Monderer wrote:
> Hi Konrad,
> 
> I've encountered that problem when I passed the interface to the guest.
> Upgrading the kernel solved the problem.

Excellent.
> But know I try to run pass the interface to the igb_uio driver (intel DPDK driver) and I get the crash.

What kind of crash?

And please don't top-post.
> 
> Regards,
> Samuel
> ________________________________________
> From: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
> Sent: Friday, April 11, 2014 8:10 PM
> To: Samuel Monderer
> Cc: xen-devel@lists.xen.org
> Subject: Re: [Xen-devel] BUG: unable to handle kernel paging request
> 
> On Thu, Apr 10, 2014 at 06:21:52AM +0000, Samuel Monderer wrote:
> > Hi,
> >
> > When trying to bind VF to igb_uio on a PV machine I get the following error the module crashes
> >       BUG: unable to handle kernel paging request Xen dmesg shows mm.c:618:d1 Could not get page ref for pfn df300 I'm new to dpdk so if you need any other information let me know.
> >
> > Regards,
> > Samuel
> >
> > Environment:
> >       Dom0:
> >               root@Smart:~# uname -a
> >                       Linux Smart 3.13-1-amd64 #1 SMP Debian 3.13.7-1 (2014-03-25) x86_64 GNU/Linux
> >       DomU:
> >                 root@vm1:~# uname -a
> >                       Linux vm1 3.13-1-amd64 #1 SMP Debian 3.13.7-1 (2014-03-25) x86_64 GNU/Linux
> >       Xen 4.3
> >       Intel X540 Ethernet controller
> >
> > root@Smart:~# lspci
> > 01:00.0 Ethernet controller: Intel Corporation Ethernet Controller 10-Gigabit X540-AT2 (rev 01)
> > 01:00.1 Ethernet controller: Intel Corporation Ethernet Controller 10-Gigabit X540-AT2 (rev 01)
> > 02:10.0 Ethernet controller: Intel Corporation X540 Ethernet Controller Virtual Function (rev 01)
> > 02:10.1 Ethernet controller: Intel Corporation X540 Ethernet Controller Virtual Function (rev 01)
> > 04:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller (rev 06) root@Smart:~#
> >
> > root@vm1:~# ./dpdk-1.6.0r1/tools/pci_unbind.py --status
> >
> > Network devices using IGB_UIO driver
> > ====================================
> > <none>
> >
> > Network devices using kernel driver
> > ===================================
> > 0000:00:00.0 'X540 Ethernet Controller Virtual Function' if=eth1 drv=ixgbevf unused=igb_uio
> >
> > Other network devices
> > =====================
> > <none>
> > root@vm1:~# ./dpdk-1.6.0r1/tools/pci_unbind.py -b igb_uio 00:00.0
> > [   90.348487] xen:events: xen_bind_pirq_gsi_to_irq: returning irq 47 for gsi 0
> > [   90.348496] igb_uio 0000:00:00.0: Xen PCI mapped GSI0 to IRQ47
> > [   90.348791] BUG: unable to handle kernel paging request at ffffc9000063e00c
> > [   90.348796] IP: [<ffffffffa01542cf>] igbuio_set_interrupt_mask.isra.2+0xbf/0xe0 [igb_uio]
> > [   90.348802] PGD 98440067 PUD 98441067 PMD 97e47067 PTE 8010000000000464
> > [   90.348807] Oops: 0003 [#1] SMP
> > [   90.348810] Modules linked in: igb_uio(O) uio ixgbevf xen_pcifront x86_pkg_temp_thermal thermal_sys coretemp crct10dif_pclmul crct10dif_common crc32_pclmul crc32c_intel ghash_clmulni_intel evdev aesni_intel aes_x86_64 lrw gf128mul glue_helper ablk_helper cryptd pcspkr ext4 crc16 mbcache jbd2 dm_mod xen_netfront xen_blkfront
> > [   90.348831] CPU: 0 PID: 1378 Comm: pci_unbind.py Tainted: G           O 3.13-1-amd64 #1 Debian 3.13.7-1
> > [   90.348835] task: ffff880094a5a010 ti: ffff880015c1a000 task.ti: ffff880015c1a000
> > [   90.348838] RIP: e030:[<ffffffffa01542cf>]  [<ffffffffa01542cf>] igbuio_set_interrupt_mask.isra.2+0xbf/0xe0 [igb_uio]
> > [   90.348843] RSP: e02b:ffff880015c1bd78  EFLAGS: 00010086
> > [   90.348845] RAX: ffff8800158f1480 RBX: 0000000000000000 RCX: 0000000000000001
> > [   90.348848] RDX: ffff8800158f1498 RSI: ffffc9000063e00c RDI: ffff880094ec1868
> > [   90.348850] RBP: ffff880094ec1000 R08: 0000000000000001 R09: 0000000000000000
> > [   90.348853] R10: 0000000000000030 R11: 0000000000000044 R12: ffff880094ec1000
> > [   90.348855] R13: 0000000000000200 R14: 0000000000000000 R15: 0000000000000002
> > [   90.348860] FS:  00007f0dfabfd700(0000) GS:ffff88009f800000(0000) knlGS:0000000000000000
> > [   90.348863] CS:  e033 DS: 0000 ES: 0000 CR0: 0000000080050033
> > [   90.348865] CR2: ffffc9000063e00c CR3: 0000000094812000 CR4: 0000000000042660
> > [   90.348868] Stack:
> > [   90.348870]  ffffffff8129861c ffff880015cc9000 ffff880015cc9218 ffffffffa0154418
> > [   90.348874]  ffff880094ec1098 ffff880094ec10a8 ffff880015cc9000 ffff880094ec1000
> > [   90.348878]  0000000000000000 ffffffffa015470c 0000000000004000 0000000000000200
> > [   90.348882] Call Trace:
> > [   90.348888]  [<ffffffff8129861c>] ? pci_cfg_access_trylock+0x3c/0x50
> > [   90.348892]  [<ffffffffa0154418>] ? igbuio_pci_irqcontrol+0x48/0x90 [igb_uio]
> > [   90.348895]  [<ffffffffa015470c>] ? igbuio_pci_probe+0x2ac/0x450 [igb_uio]
> > [   90.348899]  [<ffffffff812a108a>] ? local_pci_probe+0x3a/0xa0
> > [   90.348903]  [<ffffffff812a239a>] ? pci_device_probe+0xca/0x120
> > [   90.348908]  [<ffffffff81355058>] ? driver_probe_device+0x68/0x220
> > [   90.348911]  [<ffffffff81353b4b>] ? bind_store+0xcb/0x110
> > [   90.348916]  [<ffffffff811e74a6>] ? sysfs_write_file+0x126/0x1c0
> > [   90.348921]  [<ffffffff8117957a>] ? vfs_write+0xaa/0x1e0
> > [   90.348924]  [<ffffffff81179f83>] ? SyS_write+0x43/0xa0
> > [   90.348929]  [<ffffffff814a71c8>] ? page_fault+0x28/0x30
> > [   90.348933]  [<ffffffff814ae4f9>] ? system_call_fastpath+0x16/0x1b
> > [   90.348935] Code: 0f b7 48 02 89 f1 41 89 f0 83 e1 fe 41 83 c8 01 85 db 41 0f 44 c8 39 ce 74 1c 41 0f b7 f1 c1 e6 04 83 c6 0c 48 63 f6 48 03 70 28 <89> 0e 48 8b 50 28 8b 12 89 48 08 48 8b 48 18 48 39 cf 48 8d 41
> > [   90.348964] RIP  [<ffffffffa01542cf>] igbuio_set_interrupt_mask.isra.2+0xbf/0xe0 [igb_uio]
> 
> 
> That could be:
> https://bugzilla.kernel.org/show_bug.cgi?id=64581
> 
> which is that you need these two:
> 
> commit ac8344c4c0bf74c7efaf962cf2a6404331678ce4
> Author: DuanZhenzhong <zhenzhong.duan@oracle.com>
> Date:   Wed Dec 4 13:09:16 2013 +0800
> 
>     PCI: Drop "irq" param from *_restore_msi_irqs()
> commit 0e4ccb1505a9e29c50170742ce26ac4655baab2d
> Author: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
> Date:   Wed Nov 6 16:16:56 2013 -0500
> 
>     PCI: Add x86_msi.msi_mask_irq() and msix_mask_irq()
>