PCI-Passthrough Error: transmit queue timed out

PCI-Passthrough Error: transmit queue timed out

[openlui]

[-- Attachment #1.1: Type: text/plain, Size: 6474 bytes --]

Hi, all:
    I have tried PCI-Passthrough to DomU in Xen. However, If we send packets to DomU for a while, there is chance that the networking of Domu will be disconnected. The corresponding syslog messages are as shown at the end of mail.
   And I found the following analysis at [1], I am wondering if it is the reason for it.  Does someone also meet this problem, and can give me some advice about it?  The environment is as follows:
  Dom0: SUSE 12 (3.12.28)
  DomU: 3.17.4 mainline


[1] http://martinbj2008.github.io/blog/2014/09/17/netdevice-watchdog-cause-tx-queue-schedule/


2015-01-31T04:53:45.005161-05:00 linux-gbze kernel: [63331.820079] ------------[ cut here ]------------
2015-01-31T04:53:45.005187-05:00 linux-gbze kernel: [63331.820163] WARNING: CPU: 0 PID: 0 at net/sched/sch_generic.c:264 dev_watchdog+0x266/0x270()
2015-01-31T04:53:45.005190-05:00 linux-gbze kernel: [63331.820196] NETDEV WATCHDOG: eth0 (ixgbe): transmit queue 1 timed out
2015-01-31T04:53:45.005192-05:00 linux-gbze kernel: [63331.820227] Modules linked in: iptable_filter(E) ip_tables(E) x_tables(E) openvswitch(E) gre(E) vxlan(E) udp_tunnel(E) libcrc32c(E) xen_netback(OE) ixgbe(OE) nls_utf8(E) isofs(E) fuse(E) iscsi_ibft(E) iscsi_boot_sysfs(E) af_packet(E) intel_rapl(E) joydev(E) ppdev(E) crct10dif_pclmul(E) parport_pc(E) crc32_pclmul(E) parport(E) crc32c_intel(E) ghash_clmulni_intel(E) ptp(E) aesni_intel(E) pps_core(E) aes_x86_64(E) mdio(E) processor(E) lrw(E) dca(E) gf128mul(E) glue_helper(E) serio_raw(E) i2c_piix4(E) pcspkr(E) ablk_helper(E) button(E) cryptd(E) dm_mod(E) ext4(E) crc16(E) mbcache(E) jbd2(E) sr_mod(E) cdrom(E) ata_generic(E) ata_piix(E) ahci(E) libahci(E) xen_blkfront(E) floppy(E) cirrus(E) drm_kms_helper(E) ttm(E) drm(E) libata(E) sg(E) scsi_mod(E) autofs4(E) [last unloaded: ixgbe]


2015-01-31T04:53:45.005195-05:00 linux-gbze kernel: [63331.820337] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G           OE  3.17.4-4-default #10
2015-01-31T04:53:45.005197-05:00 linux-gbze kernel: [63331.820368] Hardware name: Xen HVM domU, BIOS 4.4.1_06-2.2 10/08/2014
2015-01-31T04:53:45.005199-05:00 linux-gbze kernel: [63331.820398]  0000000000000009 ffff88010f803db0 ffffffff8158c8e7 ffff88010f803df8
2015-01-31T04:53:45.005200-05:00 linux-gbze kernel: [63331.820432]  ffff88010f803de8 ffffffff8106d9ed 0000000000000001 ffff8800ead60000
2015-01-31T04:53:45.005202-05:00 linux-gbze kernel: [63331.820464]  0000000000000040 0000000000000000 ffff8800ead60000 ffff88010f803e48


2015-01-31T04:53:45.005203-05:00 linux-gbze kernel: [63331.820497] Call Trace:


2015-01-31T04:53:45.005205-05:00 linux-gbze kernel: [63331.820528]  <IRQ>  [<ffffffff8158c8e7>] dump_stack+0x45/0x56
2015-01-31T04:53:45.005206-05:00 linux-gbze kernel: [63331.820571]  [<ffffffff8106d9ed>] warn_slowpath_common+0x7d/0xa0
2015-01-31T04:53:45.005208-05:00 linux-gbze kernel: [63331.820603]  [<ffffffff8106da5c>] warn_slowpath_fmt+0x4c/0x50
2015-01-31T04:53:45.005210-05:00 linux-gbze kernel: [63331.820641]  [<ffffffff8100a4ef>] ? xen_timer_interrupt+0x10f/0x150
2015-01-31T04:53:45.005212-05:00 linux-gbze kernel: [63331.820675]  [<ffffffff814be3d6>] dev_watchdog+0x266/0x270
2015-01-31T04:53:45.005213-05:00 linux-gbze kernel: [63331.820708]  [<ffffffff814be170>] ? dev_graft_qdisc+0x80/0x80
2015-01-31T04:53:45.005215-05:00 linux-gbze kernel: [63331.820744]  [<ffffffff810cd656>] call_timer_fn+0x36/0x100
2015-01-31T04:53:45.005216-05:00 linux-gbze kernel: [63331.820777]  [<ffffffff814be170>] ? dev_graft_qdisc+0x80/0x80
2015-01-31T04:53:45.005218-05:00 linux-gbze kernel: [63331.820811]  [<ffffffff810ced7a>] run_timer_softirq+0x1fa/0x2e0
2015-01-31T04:53:45.005220-05:00 linux-gbze kernel: [63331.820845]  [<ffffffff81071885>] __do_softirq+0xe5/0x280
2015-01-31T04:53:45.005221-05:00 linux-gbze kernel: [63331.820878]  [<ffffffff81071c6d>] irq_exit+0xad/0xc0
2015-01-31T04:53:45.005223-05:00 linux-gbze kernel: [63331.820912]  [<ffffffff813a6dc8>] xen_evtchn_do_upcall+0x38/0x50
2015-01-31T04:53:45.005224-05:00 linux-gbze kernel: [63331.820948]  [<ffffffff81594a6d>] xen_hvm_callback_vector+0x6d/0x80
2015-01-31T04:53:45.005225-05:00 linux-gbze kernel: [63331.820978]  <EOI>  [<ffffffff81054d06>] ? native_safe_halt+0x6/0x10
2015-01-31T04:53:45.005227-05:00 linux-gbze kernel: [63331.821020]  [<ffffffff8101c7ef>] default_idle+0x1f/0xc0
2015-01-31T04:53:45.005229-05:00 linux-gbze kernel: [63331.821054]  [<ffffffff8101d15f>] arch_cpu_idle+0xf/0x20
2015-01-31T04:53:45.005231-05:00 linux-gbze kernel: [63331.821087]  [<ffffffff810ab714>] cpu_startup_entry+0x2f4/0x330
2015-01-31T04:53:45.005232-05:00 linux-gbze kernel: [63331.821121]  [<ffffffff8157cf77>] rest_init+0x77/0x80
2015-01-31T04:53:45.005234-05:00 linux-gbze kernel: [63331.821156]  [<ffffffff81b6a0c4>] start_kernel+0x46f/0x47a
2015-01-31T04:53:45.005235-05:00 linux-gbze kernel: [63331.821188]  [<ffffffff81b69a4e>] ? set_init_arg+0x53/0x53
2015-01-31T04:53:45.005236-05:00 linux-gbze kernel: [63331.821222]  [<ffffffff81b69120>] ? early_idt_handlers+0x120/0x120
2015-01-31T04:53:45.005238-05:00 linux-gbze kernel: [63331.821255]  [<ffffffff81b695ee>] x86_64_start_reservations+0x2a/0x2c
2015-01-31T04:53:45.005240-05:00 linux-gbze kernel: [63331.821288]  [<ffffffff81b69733>] x86_64_start_kernel+0x143/0x152
2015-01-31T04:53:45.005242-05:00 linux-gbze kernel: [63331.821320] ---[ end trace a40fbdbc6585a982 ]---
2015-01-31T04:53:45.005243-05:00 linux-gbze kernel: [63331.821356] ixgbe 0000:00:04.0 eth0: initiating reset due to tx timeout
2015-01-31T04:53:45.011716-05:00 linux-gbze kernel: [63331.825409] ixgbe 0000:00:04.0 eth0: Reset adapter
2015-01-31T04:53:45.011724-05:00 linux-gbze kernel: [63331.825524] ixgbe 0000:00:04.0 eth0: RXDCTL.ENABLE on Rx queue 0 not cleared within the polling period
2015-01-31T04:53:45.011725-05:00 linux-gbze kernel: [63331.825634] ixgbe 0000:00:04.0 eth0: RXDCTL.ENABLE on Rx queue 1 not cleared within the polling period
2015-01-31T04:53:45.011726-05:00 linux-gbze kernel: [63331.825744] ixgbe 0000:00:04.0 eth0: RXDCTL.ENABLE on Rx queue 2 not cleared within the polling period
2015-01-31T04:53:45.011726-05:00 linux-gbze kernel: [63331.825854] ixgbe 0000:00:04.0 eth0: RXDCTL.ENABLE on Rx queue 3 not cleared within the polling period
2015-01-31T04:53:45.407723-05:00 linux-gbze kernel: [63332.220428] ixgbe 0000:00:04.0 eth0: detected SFP+: 3
2015-01-31T04:53:45.751765-05:00 linux-gbze kernel: [63332.564118] ixgbe 0000:00:04.0 eth0: NIC Link is Up 10 Gbps, Flow Control: RX/TX

[-- Attachment #1.2: Type: text/html, Size: 7689 bytes --]

[-- Attachment #2: Type: text/plain, Size: 126 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

Re: PCI-Passthrough Error: transmit queue timed out

[Jan Beulich]

>>> On 31.01.15 at 09:49, <openlui@126.com> wrote:
> 2015-01-31T04:53:45.005161-05:00 linux-gbze kernel: [63331.820079] ------------[ cut here ]------------
> 2015-01-31T04:53:45.005187-05:00 linux-gbze kernel: [63331.820163] WARNING: CPU: 
> 0 PID: 0 at net/sched/sch_generic.c:264 dev_watchdog+0x266/0x270()
> 2015-01-31T04:53:45.005190-05:00 linux-gbze kernel: [63331.820196] NETDEV 
> WATCHDOG: eth0 (ixgbe): transmit queue 1 timed out

Please keep in mind that Intel officially doesn't support passing
through physical functions of NICs to guests; only SR-IOV VF
pass-through is supported from what they've been telling us. This
may well have to do with (wild guess!) them knowing their PF
drivers not working reliably in unprivileged guests - in any event
your report seems to hint towards a driver issue.

Jan