Realtek RTL8111/8168B IO Page Fault

[asylum]

Hello,

I have a home server running Debian Wheezy ( 3.2.0-4-amd64 #1 SMP Debian 3.2.68-1+deb7u6 x86_64 GNU/Linux). I have a Realtek RTL8111/8168B NIC and an Intel 82574L. I saw while searching Google that both cards have this issue. I'm currently using the Realtek card for LAN connectivity. If I try to tranfer large files via SAMBA or have any kind of load on the NIC, my system freezes and requires a reboot in order to restore connectivity. The error message from kern.log is:

Nov 26 02:41:53 golgotha kernel: [ 5178.413329] AMD-Vi: Event logged [IO_PAGE_FAULT device=0a:00.0 domain=0x0021 address=0x0000000000003000 flags=0x0050]
Nov 26 02:42:04 golgotha kernel: [ 5188.800088] ------------[ cut here ]------------
Nov 26 02:42:04 golgotha kernel: [ 5188.800100] WARNING: at /build/linux-Tvajqd/linux-3.2.68/net/sched/sch_generic.c:256 dev_watchdog+0xf2/0x151()
Nov 26 02:42:04 golgotha kernel: [ 5188.800105] Hardware name: To be filled by O.E.M.
Nov 26 02:42:04 golgotha kernel: [ 5188.800108] NETDEV WATCHDOG: eth0 (r8169): transmit queue 0 timed out
Nov 26 02:42:04 golgotha kernel: [ 5188.800111] Modules linked in: des_generic ecb md4 hmac cifs fscache ipt_REJECT xt_tcpudp xt_state iptable_filter iptable_mangle iptable_nat nf_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_conntrack ip_tables x_tables nls_utf8 nls_cp437 vfat fat snd_hda_codec_hdmi snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep radeon ttm drm_kms_helper snd_pcm powernow_k8 drm snd_page_alloc snd_timer mperf crc32c_intel ghash_clmulni_intel aesni_intel aes_x86_64 power_supply eeepc_wmi asus_wmi i2c_algo_bit sparse_keymap rfkill sp5100_tco mxm_wmi snd edac_mce_amd psmouse i2c_piix4 i2c_core serio_raw processor aes_generic wmi soundcore edac_core pcspkr joydev fam15h_power evdev k10temp thermal_sys cryptd button ext4 crc16 jbd2 mbcache ses dm_mod enclosure hid_logitech_dj sg usbhid sr_mod hid sd_mod cdrom crc_t10dif usb_storage ohci_hcd ahci libahci libata xhci_hcd ehci_hcd r8169 mii scsi_mod e1000e usbcore usb_common [last unloaded: scsi_wait_scan]
Nov 26 02:42:04 golgotha kernel: [ 5188.800215] Pid: 0, comm: swapper/2 Not tainted 3.2.0-4-amd64 #1 Debian 3.2.68-1+deb7u6
Nov 26 02:42:04 golgotha kernel: [ 5188.800218] Call Trace:
Nov 26 02:42:04 golgotha kernel: [ 5188.800221] <IRQ> [<ffffffff81046de1>] ? warn_slowpath_common+0x78/0x8c
Nov 26 02:42:04 golgotha kernel: [ 5188.800237] [<ffffffff81046e8d>] ? warn_slowpath_fmt+0x45/0x4a
Nov 26 02:42:04 golgotha kernel: [ 5188.800244] [<ffffffff812a8d51>] ? netif_tx_lock+0x40/0x75
Nov 26 02:42:04 golgotha kernel: [ 5188.800254] [<ffffffff812a8ec1>] ? dev_watchdog+0xf2/0x151
Nov 26 02:42:04 golgotha kernel: [ 5188.800261] [<ffffffff81052618>] ? run_timer_softirq+0x19a/0x261
Nov 26 02:42:04 golgotha kernel: [ 5188.800265] [<ffffffff812a8dcf>] ? netif_tx_unlock+0x49/0x49
Nov 26 02:42:04 golgotha kernel: [ 5188.800272] [<ffffffff81066439>] ? timekeeping_get_ns+0xd/0x2a
Nov 26 02:42:04 golgotha kernel: [ 5188.800277] [<ffffffff8104c48e>] ? __do_softirq+0xb9/0x177
Nov 26 02:42:04 golgotha kernel: [ 5188.800284] [<ffffffff813584ec>] ? call_softirq+0x1c/0x30
Nov 26 02:42:04 golgotha kernel: [ 5188.800291] [<ffffffff8100fa91>] ? do_softirq+0x3c/0x7b
Nov 26 02:42:04 golgotha kernel: [ 5188.800295] [<ffffffff8104c6f6>] ? irq_exit+0x3c/0x99
Nov 26 02:42:04 golgotha kernel: [ 5188.800302] [<ffffffff8102475c>] ? smp_apic_timer_interrupt+0x74/0x82
Nov 26 02:42:04 golgotha kernel: [ 5188.800308] [<ffffffff81356d5e>] ? apic_timer_interrupt+0x6e/0x80
Nov 26 02:42:04 golgotha kernel: [ 5188.800311] <EOI> [<ffffffff810623ec>] ? enqueue_hrtimer+0x43/0x67
Nov 26 02:42:04 golgotha kernel: [ 5188.800329] [<ffffffffa01f6358>] ? arch_local_irq_enable+0x4/0x8 [processor]
Nov 26 02:42:04 golgotha kernel: [ 5188.800341] [<ffffffffa01f6fea>] ? acpi_idle_enter_simple+0xc6/0x102 [processor]
Nov 26 02:42:04 golgotha kernel: [ 5188.800351] [<ffffffff81271b9d>] ? cpuidle_idle_call+0xec/0x179
Nov 26 02:42:04 golgotha kernel: [ 5188.800357] [<ffffffff8100d242>] ? cpu_idle+0xa5/0xf2
Nov 26 02:42:04 golgotha kernel: [ 5188.800364] [<ffffffff8134328a>] ? start_secondary+0x1da/0x1e0
Nov 26 02:42:04 golgotha kernel: [ 5188.800368] ---[ end trace fbbea9ad07586541 ]---

I've scoured Google but have been unable to fix this issue. Everything runs fine until I try to do a large file tranfer or have multiple users connected. I have seen many posts with similar symptoms, but none of the posted fixes have helped.

lspci returns:

09:00.0 Ethernet controller: Intel Corporation 82574L Gigabit Network Connection
0a:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 09)

Also, I've rolled the driver back to rtl_nic/rtl8168f-1.fw and currently have that installed, but the error still occurs.

Any help or advice is greatly appreciated. Thanks!

[v&n]

I have seen many posts with similar symptoms, but none of the posted fixes have helped.

It is always a good idea to post all (or as many as you can remember) the links that you actually tried. Unless you started fresh after each try.

lspci returns:

09:00.0 Ethernet controller: Intel Corporation 82574L Gigabit Network Connection
0a:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 09)

Also, I've rolled the driver back to rtl_nic/rtl8168f-1.fw and currently have that installed, but the error still occurs.

That is a firmware, not driver. Can we see which realtek driver you currently are working with? That is -

Code: Select all

lspci -nnk | grep -iA2 net
lsmod

You may strip off irrelevant part from the output of lsmod, but post the full output if are not sure.

The recommended driver for your card is r8168 (the one in kernel is r8169). There is a dkms package available for that, unfortunately only for sid and stretch. So you may have to compile it yourself (a Github source : https://github.com/mtorromeo/r8168). You may try the dkms package anyway, but think I tried it once on wheezy and it failed to build as expected.

[asylum]

Hi v&n,

From lspci it appears that the r8169 module is being used again. I had it blacklisted, but a kernel update may have overwritten that. A few of the posts that I have seen are:

http://ubuntuforums.org/showthread.php?t=1983040
https://bugzilla.kernel.org/show_bug.cgi?id=14962#c26

And this is the post that I used to load the r8168 driver:

https://unixblogger.wordpress.com/2011/ ... rnet-card/


root@golgotha /var/log# lspci -nnk | grep -iA2 net
09:00.0 Ethernet controller [0200]: Intel Corporation 82574L Gigabit Network Connection [8086:10d3]
Subsystem: Intel Corporation Gigabit CT Desktop Adapter [8086:a01f]
Kernel driver in use: e1000e
0a:00.0 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller [10ec:8168] (rev 09)
Subsystem: ASUSTeK Computer Inc. Device [1043:8505]
Kernel driver in use: r8169

root@golgotha /var/log# lsmod
Module Size Used by
ses 13106 0
enclosure 13306 1 ses
usb_storage 43870 0
des_generic 20851 0
ecb 12737 0
md4 12529 0
hmac 12835 1
nls_utf8 12456 1
cifs 243028 2
fscache 36739 1 cifs
ipt_REJECT 12502 2
xt_tcpudp 12570 88
xt_state 12503 1
iptable_filter 12536 1
iptable_mangle 12536 0
iptable_nat 12928 0
nf_nat 18242 1 iptable_nat
nf_conntrack_ipv4 14078 4 nf_nat,iptable_nat
nf_defrag_ipv4 12483 1 nf_conntrack_ipv4
nf_conntrack 52720 4 nf_conntrack_ipv4,nf_nat,iptable_nat,xt_state
ip_tables 22042 3 iptable_nat,iptable_mangle,iptable_filter
x_tables 19118 7 ip_tables,iptable_nat,iptable_mangle,iptable_filter,xt_state,xt_tcpudp,ipt_REJECT
vfat 17316 0
fat 45642 1 vfat
snd_hda_codec_hdmi 30824 1
powernow_k8 17618 0
mperf 12453 1 powernow_k8
crc32c_intel 12747 0
ghash_clmulni_intel 13130 0
snd_hda_codec_realtek 188851 1
aesni_intel 50667 0
radeon 722295 1
aes_x86_64 16843 1 aesni_intel
sp5100_tco 12900 0
eeepc_wmi 12564 0
snd_hda_intel 26259 0
psmouse 69265 0
snd_hda_codec 78031 3 snd_hda_intel,snd_hda_codec_realtek,snd_hda_codec_hdmi
snd_hwdep 13186 1 snd_hda_codec
snd_pcm 68083 3 snd_hda_codec,snd_hda_intel,snd_hda_codec_hdmi
snd_page_alloc 13003 2 snd_pcm,snd_hda_intel
snd_timer 22917 1 snd_pcm
asus_wmi 18726 1 eeepc_wmi
edac_mce_amd 17103 0
aes_generic 33026 2 aes_x86_64,aesni_intel
sparse_keymap 12760 1 asus_wmi
rfkill 19012 1 asus_wmi
ttm 53664 1 radeon
drm_kms_helper 31370 1 radeon
drm 183952 3 drm_kms_helper,ttm,radeon
i2c_piix4 12536 0
power_supply 13475 1 radeon
i2c_algo_bit 12841 1 radeon
processor 28149 7 powernow_k8
mxm_wmi 12515 0
i2c_core 23876 5 i2c_algo_bit,i2c_piix4,drm,drm_kms_helper,radeon
serio_raw 12931 0
snd 52893 7 snd_pcm,snd_timer,snd_hwdep,snd_hda_codec,snd_hda_intel,snd_hda_codec_realtek,snd_hda_codec_hdmi
edac_core 35258 0
pcspkr 12579 0
cryptd 14517 2 aesni_intel,ghash_clmulni_intel
thermal_sys 18040 1 processor
fam15h_power 12677 0
button 12937 0
wmi 13243 2 mxm_wmi,asus_wmi
k10temp 12611 0
joydev 17266 0
evdev 17562 6
soundcore 13065 1 snd
ext4 350804 5
crc16 12343 1 ext4
jbd2 62115 1 ext4
mbcache 13114 1 ext4
dm_mod 63645 0
sr_mod 21899 0
hid_logitech_dj 17313 0
usbhid 36418 1 hid_logitech_dj
hid 81372 2 usbhid,hid_logitech_dj
sg 25874 0
cdrom 35401 1 sr_mod
sd_mod 36136 10
crc_t10dif 12348 1 sd_mod
ohci_hcd 26563 0
e1000e 133833 0
xhci_hcd 73352 0
ahci 24997 6
libahci 22941 1 ahci
ehci_hcd 40249 0
r8169 47008 0
mii 12675 1 r8169
libata 140630 2 libahci,ahci
usbcore 128741 6 ehci_hcd,xhci_hcd,ohci_hcd,usbhid,usb_storage
scsi_mod 162321 6 libata,sd_mod,sg,sr_mod,usb_storage,ses
usb_common 12354 1 usbcore


I was able to tranfer a file just now, less that a gig though after installing updates. Not sure if anything is fixed or just wishful thinking. If the issue is still there I should know fairly quickly though. Thanks!

[asylum]

Spoke too soon, just crashed again. Looks like I may have to try rolling back to the r8168 driver again. X) Thanks!

[v&n]

Yes. The native driver (r8169) worked quite well for a quite long time (between kernels 3.5 to 3.11 or even later), but its older problems seem to have been resurrected in more recent kernels.

By the way, you should always use 'Code' tags (click on the 'Code' button above the edit box > copy-past the output in-between the generated 'code' tags pair) to post terminal commands and their outputs. It preserves the formatting, making it more readable.

[asylum]

Thanks v&n. That explains why I see posts from 2010 - 2011, then nothing, and the more posts from 2014 - present. I'll give it a shot with the r8168 again and post how it goes.