Bug#992131: R8169 CRASH! (rtl_rxtx_empty_cond == 0 (loop: 42, delay: 100))

[Vincenzo Gianfelice]

Package: r8169

Kernel: 5.10.0-8 (amd64)

System: Debian 11 (bullseye)

libc: 2.31

Hi. I state that this problem is present from kernel 5.4 and still has not been fixed ... suddenly, the ethernet stops working and this error appears (in dmesg):

[   12.063469] NETDEV WATCHDOG: eth0 (r8169): transmit queue 0 timed out

[   12.063509] WARNING: CPU: 3 PID: 0 at net/sched/sch_generic.c:467 dev_watchdog+0x24d/0x260

[   12.063520] Modules linked in: joydev hid_logitech_hidpp uvcvideo videobuf2_vmalloc hid_gener
ic videobuf2_memops videobuf2_v4l2 videobuf2_common snd_usb_audio videodev snd_usbmidi_lib snd_r
awmidi usbhid snd_seq_device mc hid intel_rapl_msr intel_rapl_common rfkill snd_hda_codec_realte
k snd_hda_codec_generic ledtrig_audio snd_hda_codec_hdmi snd_hda_intel snd_intel_dspcfg x86_pkg_
temp_thermal soundwire_intel intel_powerclamp soundwire_generic_allocation snd_soc_core coretemp
snd_compress soundwire_cadence kvm_intel snd_hda_codec snd_hda_core nvidia_drm(POE) mei_hdcp kvm
snd_hwdep soundwire_bus irqbypass drm_kms_helper snd_pcm rapl intel_cstate snd_timer iTCO_wdt m
ei_me snd intel_pmc_bxt iTCO_vendor_support intel_uncore mei watchdog soundcore ee1004 sg intel_
wmi_thunderbolt pcspkr cec nvidia_modeset(POE) evdev acpi_pad intel_pmc_core nvidia(POE) corefre
qk(OE) parport_pc ppdev drm lp parport fuse configfs ip_tables x_tables autofs4 ext4 crc16 mbcac
he jbd2 crc32c_generic uas usb_storage sd_mod

[   12.063805]  t10_pi crc_t10dif crct10dif_generic crct10dif_pclmul crct10dif_common crc32_pclm
ul crc32c_intel ahci libahci ghash_clmulni_intel libata xhci_pci scsi_mod xhci_hcd aesni_intel i
2c_i801 i2c_smbus usbcore r8169 libaes crypto_simd realtek mdio_devres cryptd libphy glue_helper
usb_common wmi fan video button

[   12.063917] CPU: 3 PID: 0 Comm: swapper/3 Tainted: P           OE     5.10.0-8-amd64 #1 Debia
n 5.10.46-4

[   12.063922] Hardware name: Gigabyte Technology Co., Ltd. H110M-S2H/H110M-S2H-CF, BIOS F24 12/
14/2017

[   12.063933] RIP: 0010:dev_watchdog+0x24d/0x260

[   12.063941] Code: 29 cb fd ff eb a9 4c 89 f7 c6 05 72 b7 10 01 01 e8 b8 a0 fa ff 44 89 e9 4c 
89 f6 48 c7 c7 40 8d d6 93 48 89 c2 e8 d6 24 14 00 <0f> 0b eb 8a 66 66 2e 0f 1f 84 00 00 00 00 0
0 0f 1f 40 00 0f 1f 44

[   12.063947] RSP: 0018:ffffa1590018ceb8 EFLAGS: 00010286

[   12.063957] RAX: 0000000000000000 RBX: ffff8eea4c5ba800 RCX: 000000000000083f

[   12.063963] RDX: 0000000000000000 RSI: 00000000000000f6 RDI: 000000000000083f

[   12.063972] RBP: ffff8eea4b5043dc R08: 0000000000000000 R09: ffffa1590018ccd8

[   12.063978] R10: ffffa1590018ccd0 R11: ffffffff942cb3e8 R12: ffff8eea4b504480

[   12.063983] R13: 0000000000000000 R14: ffff8eea4b504000 R15: ffff8eea4c5ba880

[   12.063990] FS:  0000000000000000(0000) GS:ffff8eecaecc0000(0000) knlGS:0000000000000000

[   12.063995] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033

[   12.064001] CR2: 00005624fe890068 CR3: 0000000103f06006 CR4: 00000000003706e0

[   12.064006] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000

[   12.064011] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400

[   12.064016] Call Trace:

[   12.064024]  <IRQ>

[   12.064038]  ? pfifo_fast_enqueue+0x150/0x150

[   12.064056]  call_timer_fn+0x29/0xf0

[   12.064063]  __run_timers.part.0+0x1d3/0x240

[   12.064071]  ? ktime_get+0x38/0xa0

[   12.064079]  ? lapic_next_deadline+0x28/0x30

[   12.064087]  ? clockevents_program_event+0x8d/0xf0

[   12.064094]  run_timer_softirq+0x26/0x50

[   12.064103]  __do_softirq+0xc5/0x275

[   12.064111]  asm_call_irq_on_stack+0xf/0x20

[   12.064117]  </IRQ>

[   12.064126]  do_softirq_own_stack+0x37/0x40

[   12.064136]  irq_exit_rcu+0x8e/0xc0

[   12.064146]  sysvec_apic_timer_interrupt+0x36/0x80

[   12.064156]  asm_sysvec_apic_timer_interrupt+0x12/0x2
0

[   12.064167] RIP: 0010:cpuidle_enter_state+0xc7/0x350

[   12.064174] Code: 8b 3d 2d 34 d7 6c e8 a8 2a a2 ff 49 89 c5 0f 1f 44 00 00 31 ff e8 39 35 a2 
ff 45 84 ff 0f 85 fa 00 00 00 fb 66 0f 1f 44 00 00 <45> 85 f6 0f 88 06 01 00 00 49 63 c6 4c 2b 2
c 24 48 8d 14 40 48 8d

[   12.064180] RSP: 0018:ffffa159000dfea8 EFLAGS: 00000246

[   12.064188] RAX: ffff8eecaecebc00 RBX: 0000000000000001 RCX: 000000000000001f

[   12.064193] RDX: 0000000000000000 RSI: 000000001fefa71c RDI: 0000000000000000

[   12.064198] RBP: ffff8eecaecf5b00 R08: 00000002cf095853 R09: 0000000000000018

[   12.064204] R10: 000000000001fc29 R11: 0000000000007786 R12: ffffffff943ae6c0

[   12.064208] R13: 00000002cf095853 R14: 0000000000000001 R15: 0000000000000000

[   12.064221]  ? cpuidle_enter_state+0xb7/0x350

[   12.064230]  cpuidle_enter+0x29/0x40

[   12.064239]  do_idle+0x1ef/0x2b0

[   12.064247]  cpu_startup_entry+0x19/0x20

[   12.064256]  secondary_startup_64_no_verify+0xb0/0xbb

[   12.064264] ---[ end trace 65cac9f8dabfed2a ]---

[   12.076220] r8169 0000:02:00.0 eth0: rtl_rxtx_empty_cond == 0 (loop: 42, delay: 100).

[   22.060256] r8169 0000:02:00.0 eth0: rtl_rxtx_empty_cond == 0 (loop: 42, delay: 100).

[   32.044049] r8169 0000:02:00.0 eth0: rtl_rxtx_empty_cond == 0 (loop: 42, delay: 100).
i have no idea what might happen to generate this error. I noticed that it happens when I restart the PC a couple of times. Many people have this problem. I can't currently use ethernet!

Thanks

[a...@prnet.info]

I have seen these lines too yesterday on Debian 11, but on Linux
5.15.0-0.bpo.3-amd64.
kernel: r8169 0000:03:00.1 eno1: rtl_txcfg_empty_cond == 0 (loop: 42,
delay: 100).
kernel: r8169 0000:03:00.1 eno1: rtl_rxtx_empty_cond == 0 (loop: 42,
delay: 100).
Out of sudden like 2 times, "nmcli network off && nmcli network on"
helped to temporarily make interface working.
No big updates recently. Only these from apt log:
https://paste.nolog.cz/?fc818de693f3fd26#ETu59xDaCm4Yeqv8PP4ch8ijdbwjaiuSLSgM2ESqWDhJ

[Vincas Dargis]

I've just started getting this on Sid with 6.3:

Jun 19 20:17:43 kernel: r8169 0000:05:00.1: can't disable ASPM; OS doesn't have ASPM control
Jun 19 20:17:43 kernel: r8169 0000:05:00.1 enp5s0f1: rtl_rxtx_empty_cond == 0 (loop: 42, delay: 100).

[Vincas Dargis]

This is probably relevant upstream bug report:

https://bugzilla.kernel.org/show_bug.cgi?id=209839

[Vincas Dargis]

Still happening with 6.6 on Sid...