Sorry, the log files seem to be missing in my previous post. I thought attachment files would be added to the post. Copying the logs below.
--------------------------------------------------------------------- sudo lspci -vvvv -s 01:00.0 01:00.0 Ethernet controller: Intel Corporation 82574L Gigabit Network Connection Subsystem: Intel Corporation Device 0000 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr+ Stepping- SERR+ FastB2B- DisINTx+ Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx- Latency: 0, Cache Line Size: 64 bytes Interrupt: pin A routed to IRQ 130 Region 0: Memory at 10000000 (32-bit, non-prefetchable) [size=128K] Region 2: I/O ports at 10000000 [size=32] Region 3: Memory at 10020000 (32-bit, non-prefetchable) [size=16K] Capabilities: [c8] Power Management version 2 Flags: PMEClk- DSI+ D1- D2- AuxCurrent=0mA PME(D0+,D1-,D2-,D3hot+,D3cold+) Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=1 PME- Capabilities: [d0] MSI: Enable- Count=1/1 Maskable- 64bit+ Address: 0000000000000000 Data: 0000 Capabilities: [e0] Express (v1) Endpoint, MSI 00 DevCap: MaxPayload 256 bytes, PhantFunc 0, Latency L0s <512ns, L1 <64us ExtTag- AttnBtn- AttnInd- PwrInd- RBE+ FLReset- DevCtl: Report errors: Correctable- Non-Fatal- Fatal- Unsupported- RlxdOrd+ ExtTag- PhantFunc- AuxPwr- NoSnoop+ MaxPayload 128 bytes, MaxReadReq 512 bytes DevSta: CorrErr+ UncorrErr- FatalErr- UnsuppReq+ AuxPwr+ TransPend- LnkCap: Port #2, Speed 2.5GT/s, Width x1, ASPM L1, Latency L0 <128ns, L1 <64us ClockPM- Surprise- LLActRep- BwNot- LnkCtl: ASPM Disabled; RCB 64 bytes Disabled- Retrain- CommClk+ ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt- LnkSta: Speed 2.5GT/s, Width x1, TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt- Capabilities: [a0] MSI-X: Enable+ Count=5 Masked- Vector table: BAR=3 offset=00000000 PBA: BAR=3 offset=00002000 Kernel driver in use: e1000e ----------------------------------------------------------------- sudo ethtool -e eth0 Offset Values ------ ------ 0x0000: 00 a0 0c 1c e1 ee 20 0c 46 f7 10 20 ff ff ff ff 0x0010: fa fa 00 00 6b 02 00 00 86 80 d3 10 ef be 58 80 0x0020: 00 00 01 20 74 7e ef be 00 10 c8 00 00 00 04 27 0x0030: c9 6c 50 31 3a 07 0b 46 84 2d 00 01 00 f0 06 07 0x0040: 00 60 80 00 04 0f ff 7f 01 4f 00 00 00 00 ff 20 0x0050: 28 00 03 00 00 00 00 00 00 00 03 00 00 00 ff ff 0x0060: 43 49 49 44 52 53 44 41 ef be ef be ef be ff ff 0x0070: ef be ef be ef be ef be ef be 00 01 ff ff f4 fd -------------------------------------------------------------------- [ 4.034935] e1000e: Intel(R) PRO/1000 Network Driver - 3.0.4-NAPI [ 4.041037] e1000e: Copyright(c) 1999 - 2014 Intel Corporation. [ 4.047051] e1000e 0000:01:00.0: Disabling ASPM L0s L1 [ 4.052317] e1000e 0000:01:00.0: Interrupt Throttling Rate (ints/sec) set to dynamic conservative mode [ 4.070553] e1000e 0000:01:00.0: (unregistered net_device): Masking off all interrupts [ 4.090602] e1000e 0000:01:00.0: (unregistered net_device): Issuing a global reset to MAC [ 4.128727] e1000e 0000:01:00.0: (unregistered net_device): Masking off all interrupts [ 4.148775] e1000e 0000:01:00.0: (unregistered net_device): Issuing a global reset to MAC [ 4.186251] e1000e 0000:01:00.0: (unregistered net_device): Initializing the IEEE VLAN [ 4.186660] e1000e 0000:01:00.0: (unregistered net_device): Programming MAC Address into RAR[0] [ 4.186678] e1000e 0000:01:00.0: (unregistered net_device): Clearing RAR[1-14] [ 4.186775] e1000e 0000:01:00.0: (unregistered net_device): Zeroing the MTA [ 4.186886] e1000e 0000:01:00.0: (unregistered net_device): After fix-ups FlowControl is now = 3 [ 4.187732] e1000e 0000:01:00.0: (unregistered net_device): Reconfiguring auto-neg advertisement params [ 4.187867] e1000e 0000:01:00.0: (unregistered net_device): autoneg_advertised 2f [ 4.187877] e1000e 0000:01:00.0: (unregistered net_device): Advertise 10mb Half duplex [ 4.187887] e1000e 0000:01:00.0: (unregistered net_device): Advertise 10mb Full duplex [ 4.187897] e1000e 0000:01:00.0: (unregistered net_device): Advertise 100mb Half duplex [ 4.187907] e1000e 0000:01:00.0: (unregistered net_device): Advertise 100mb Full duplex [ 4.187917] e1000e 0000:01:00.0: (unregistered net_device): Advertise 1000mb Full duplex [ 4.187990] e1000e 0000:01:00.0: (unregistered net_device): Auto-Neg Advertising de1 [ 4.188061] e1000e 0000:01:00.0: (unregistered net_device): Restarting Auto-Neg [ 4.189545] e1000e 0000:01:00.0: (unregistered net_device): Unable to establish link!!! [ 4.189556] e1000e 0000:01:00.0: (unregistered net_device): Initializing the Flow Control address, type and timer regs [ 4.192060] e1000e 0000:01:00.0: (unregistered net_device): Phy info is only valid if link is up [ 4.192622] e1000e 0000:01:00.0: eth0: (PCI Express:2.5GT/s:Width x1) 00:a0:0c:1c:e1:ee [ 4.200686] e1000e 0000:01:00.0: eth0: Intel(R) PRO/1000 Network Connection [ 4.207772] e1000e 0000:01:00.0: eth0: PBA string buffer too small [ 4.207784] e1000e 0000:01:00.0: eth0: MAC: 3, PHY: 8, PBA No: Unknown [ 9.027468] e1000e 0000:01:00.0: eth0: Masking off all interrupts [ 9.046243] e1000e 0000:01:00.0: eth0: Issuing a global reset to MAC [ 9.086320] e1000e 0000:01:00.0: eth0: Initializing the IEEE VLAN [ 9.086767] e1000e 0000:01:00.0: eth0: Programming MAC Address into RAR[0] [ 9.086788] e1000e 0000:01:00.0: eth0: Clearing RAR[1-14] [ 9.086890] e1000e 0000:01:00.0: eth0: Zeroing the MTA [ 9.087001] e1000e 0000:01:00.0: eth0: After fix-ups FlowControl is now = 3 [ 9.087845] e1000e 0000:01:00.0: eth0: Reconfiguring auto-neg advertisement params [ 9.087975] e1000e 0000:01:00.0: eth0: autoneg_advertised 2f [ 9.087983] e1000e 0000:01:00.0: eth0: Advertise 10mb Half duplex [ 9.087990] e1000e 0000:01:00.0: eth0: Advertise 10mb Full duplex [ 9.087997] e1000e 0000:01:00.0: eth0: Advertise 100mb Half duplex [ 9.088005] e1000e 0000:01:00.0: eth0: Advertise 100mb Full duplex [ 9.088012] e1000e 0000:01:00.0: eth0: Advertise 1000mb Full duplex [ 9.088081] e1000e 0000:01:00.0: eth0: Auto-Neg Advertising de1 [ 9.088150] e1000e 0000:01:00.0: eth0: Restarting Auto-Neg [ 9.089622] e1000e 0000:01:00.0: eth0: Unable to establish link!!! [ 9.089629] e1000e 0000:01:00.0: eth0: Initializing the Flow Control address, type and timer regs [ 9.089954] e1000e 0000:01:00.0: eth0: Phy info is only valid if link is up [ 9.090332] e1000e 0000:01:00.0: eth0: After fix-ups FlowControl is now = 3 [ 9.091184] e1000e 0000:01:00.0: eth0: Reconfiguring auto-neg advertisement params [ 9.091324] e1000e 0000:01:00.0: eth0: autoneg_advertised 2f [ 9.091336] e1000e 0000:01:00.0: eth0: Advertise 10mb Half duplex [ 9.091346] e1000e 0000:01:00.0: eth0: Advertise 10mb Full duplex [ 9.091353] e1000e 0000:01:00.0: eth0: Advertise 100mb Half duplex [ 9.091361] e1000e 0000:01:00.0: eth0: Advertise 100mb Full duplex [ 9.091368] e1000e 0000:01:00.0: eth0: Advertise 1000mb Full duplex [ 9.091438] e1000e 0000:01:00.0: eth0: Auto-Neg Advertising de1 [ 9.091507] e1000e 0000:01:00.0: eth0: Restarting Auto-Neg [ 9.093007] e1000e 0000:01:00.0: eth0: Unable to establish link!!! [ 9.093019] e1000e 0000:01:00.0: eth0: Initializing the Flow Control address, type and timer regs [ 96.666726] e1000e 0000:01:00.0: eth0: Clearing RAR[1-14] [ 96.666821] e1000e 0000:01:00.0: eth0: Zeroing the MTA [ 96.666931] e1000e 0000:01:00.0: eth0: After fix-ups FlowControl is now = 3 [ 96.667770] e1000e 0000:01:00.0: eth0: Reconfiguring auto-neg advertisement params [ 96.667903] e1000e 0000:01:00.0: eth0: autoneg_advertised 2f [ 96.667911] e1000e 0000:01:00.0: eth0: Advertise 10mb Half duplex [ 96.667920] e1000e 0000:01:00.0: eth0: Advertise 10mb Full duplex [ 96.667928] e1000e 0000:01:00.0: eth0: Advertise 100mb Half duplex [ 96.667937] e1000e 0000:01:00.0: eth0: Advertise 100mb Full duplex [ 96.667945] e1000e 0000:01:00.0: eth0: Advertise 1000mb Full duplex [ 96.668016] e1000e 0000:01:00.0: eth0: Auto-Neg Advertising de1 [ 96.668086] e1000e 0000:01:00.0: eth0: Restarting Auto-Neg [ 96.669549] e1000e 0000:01:00.0: eth0: Unable to establish link!!! [ 96.669558] e1000e 0000:01:00.0: eth0: Initializing the Flow Control address, type and timer regs [ 96.672055] e1000e 0000:01:00.0: eth0: Phy info is only valid if link is up [ 96.672319] e1000e: eth0 NIC Link is Down [ 96.791973] e1000e 0000:01:00.0: eth0: Masking off all interrupts [ 96.806276] e1000e 0000:01:00.0: eth0: Issuing a global reset to MAC [ 96.846286] e1000e 0000:01:00.0: eth0: Initializing the IEEE VLAN [ 96.846694] e1000e 0000:01:00.0: eth0: Programming MAC Address into RAR[0] [ 96.846711] e1000e 0000:01:00.0: eth0: Clearing RAR[1-14] [ 96.846806] e1000e 0000:01:00.0: eth0: Zeroing the MTA [ 96.846917] e1000e 0000:01:00.0: eth0: After fix-ups FlowControl is now = 3 [ 96.847758] e1000e 0000:01:00.0: eth0: Reconfiguring auto-neg advertisement params [ 96.847891] e1000e 0000:01:00.0: eth0: autoneg_advertised 2f [ 96.847900] e1000e 0000:01:00.0: eth0: Advertise 10mb Half duplex [ 96.847908] e1000e 0000:01:00.0: eth0: Advertise 10mb Full duplex [ 96.847917] e1000e 0000:01:00.0: eth0: Advertise 100mb Half duplex [ 96.847925] e1000e 0000:01:00.0: eth0: Advertise 100mb Full duplex [ 96.847934] e1000e 0000:01:00.0: eth0: Advertise 1000mb Full duplex [ 96.848004] e1000e 0000:01:00.0: eth0: Auto-Neg Advertising de1 [ 96.848074] e1000e 0000:01:00.0: eth0: Restarting Auto-Neg [ 96.849542] e1000e 0000:01:00.0: eth0: Unable to establish link!!! [ 96.849551] e1000e 0000:01:00.0: eth0: Initializing the Flow Control address, type and timer regs [ 96.849868] e1000e 0000:01:00.0: eth0: Phy info is only valid if link is up [ 96.850233] e1000e 0000:01:00.0: eth0: After fix-ups FlowControl is now = 3 [ 96.851071] e1000e 0000:01:00.0: eth0: Reconfiguring auto-neg advertisement params [ 96.851202] e1000e 0000:01:00.0: eth0: autoneg_advertised 2f [ 96.851210] e1000e 0000:01:00.0: eth0: Advertise 10mb Half duplex [ 96.851219] e1000e 0000:01:00.0: eth0: Advertise 10mb Full duplex [ 96.851227] e1000e 0000:01:00.0: eth0: Advertise 100mb Half duplex [ 96.851236] e1000e 0000:01:00.0: eth0: Advertise 100mb Full duplex [ 96.851244] e1000e 0000:01:00.0: eth0: Advertise 1000mb Full duplex [ 96.851313] e1000e 0000:01:00.0: eth0: Auto-Neg Advertising de1 [ 96.851383] e1000e 0000:01:00.0: eth0: Restarting Auto-Neg [ 96.852853] e1000e 0000:01:00.0: eth0: Unable to establish link!!! [ 96.852862] e1000e 0000:01:00.0: eth0: Initializing the Flow Control address, type and timer regs [ 99.886438] e1000e 0000:01:00.0: eth0: hw->fc.current_mode = 3 [ 99.886697] e1000e 0000:01:00.0: eth0: Flow Control = FULL. [ 99.886709] e1000e 0000:01:00.0: eth0: 1000 Mbps, Full Duplex [ 99.886720] e1000e 0000:01:00.0: eth0: hw->fc.current_mode = 3 [ 99.887264] e1000e 0000:01:00.0: eth0: 1000 Mbps, Full Duplex [ 99.887275] e1000e: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx [ 122.966345] e1000e: eth0 NIC Link is Down [ 125.586436] e1000e 0000:01:00.0: eth0: hw->fc.current_mode = 3 [ 125.586695] e1000e 0000:01:00.0: eth0: Flow Control = FULL. [ 125.586707] e1000e 0000:01:00.0: eth0: 1000 Mbps, Full Duplex [ 125.586719] e1000e 0000:01:00.0: eth0: hw->fc.current_mode = 3 [ 125.587267] e1000e 0000:01:00.0: eth0: 1000 Mbps, Full Duplex [ 125.587278] e1000e: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx [ 2339.326389] e1000e: eth0 NIC Link is Down [ 2343.966431] e1000e 0000:01:00.0: eth0: hw->fc.current_mode = 3 [ 2343.966689] e1000e 0000:01:00.0: eth0: Flow Control = FULL. [ 2343.966701] e1000e 0000:01:00.0: eth0: 1000 Mbps, Full Duplex [ 2343.966712] e1000e 0000:01:00.0: eth0: hw->fc.current_mode = 3 [ 2343.967259] e1000e 0000:01:00.0: eth0: 1000 Mbps, Full Duplex [ 2343.967270] e1000e: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx [ 2549.936903] CPU1: Booted secondary processor [ 2549.946191] Switched to NOHz mode on CPU #1 [ 3354.616201] ------------[ cut here ]------------ [ 3354.616231] WARNING: at /home/tony/work/git/kernel/linux-tegra/net/sched/sch_generic.c:255 dev_watchdog+0x298/0x2bc() [ 3354.616243] NETDEV WATCHDOG: eth0 (e1000e): transmit queue 0 timed out [ 3354.616251] Modules linked in: snd_hda_codec_hdmi wl18xx wlcore_sdio wlcore bnep mac80211 rfcomm cfg80211 snd_hda_intel snd_hda_codec compat btwilink [ 3354.616290] Backtrace: [ 3354.616323] [<c0011e0c>] (dump_backtrace+0x0/0x10c) from [<c058f648>] (dump_stack+0x18/0x1c) [ 3354.616333] r6:000000ff r5:c04a280c r4:c0789da8 r3:00000000 [ 3354.616358] [<c058f630>] (dump_stack+0x0/0x1c) from [<c005af88>] (warn_slowpath_common+0x54/0x6c) [ 3354.616372] [<c005af34>] (warn_slowpath_common+0x0/0x6c) from [<c005b044>] (warn_slowpath_fmt+0x38/0x40) [ 3354.616382] r8:e6b3fca0 r7:c0852754 r6:c078a080 r5:00000000 r4:e6870000 [ 3354.616398] r3:00000009 [ 3354.616410] [<c005b00c>] (warn_slowpath_fmt+0x0/0x40) from [<c04a280c>] (dev_watchdog+0x298/0x2bc) [ 3354.616420] r3:e6870000 r2:c071ac08 [ 3354.616439] [<c04a2574>] (dev_watchdog+0x0/0x2bc) from [<c0066f4c>] (run_timer_softirq+0x104/0x218) [ 3354.616459] [<c0066e48>] (run_timer_softirq+0x0/0x218) from [<c0060ea4>] (__do_softirq+0xc4/0x168) [ 3354.616475] [<c0060de0>] (__do_softirq+0x0/0x168) from [<c0061400>] (irq_exit+0x9c/0xa4) [ 3354.616490] [<c0061364>] (irq_exit+0x0/0xa4) from [<c00136dc>] (ipi_timer+0x44/0x48) [ 3354.616499] r4:c0786060 r3:00000000 [ 3354.616513] [<c0013698>] (ipi_timer+0x0/0x48) from [<c0008354>] (do_local_timer+0x5c/0x78) [ 3354.616522] r5:c07aa268 r4:c0788000 [ 3354.616537] [<c00082f8>] (do_local_timer+0x0/0x78) from [<c000de78>] (__irq_svc+0x38/0xd0) [ 3354.616547] Exception stack(0xc0789ee8 to 0xc0789f30) [ 3354.616558] 9ee0: c0789f38 00000000 00000000 000f4240 e6ae8400 e6ae8470 [ 3354.616571] 9f00: c081b25c e6ae8470 00001a87 00000000 00000000 c0789f64 3b9ac9ff c0789f30 [ 3354.616582] 9f20: c022ae5c c004256c 20000153 ffffffff [ 3354.616589] r8:00001a87 r7:c0789f1c r6:fe040100 r5:20000153 r4:c004256c [ 3354.616605] r3:c022ae5c [ 3354.616623] [<c00424b0>] (tegra_idle_enter_lp2+0x0/0x10c) from [<c0414898>] (cpuidle_idle_call+0xbc/0x148) [ 3354.616633] r9:412fc099 r8:8000406a r6:c08511c8 r5:00000000 r4:e6ae8400 [ 3354.616656] [<c04147dc>] (cpuidle_idle_call+0x0/0x148) from [<c000f288>] (cpu_idle+0xc0/0x118) [ 3354.616666] r9:412fc099 r8:8000406a r7:c07adf4c r6:c05a08c0 r5:c0817924 [ 3354.616682] r4:c0788000 [ 3354.616698] [<c000f1c8>] (cpu_idle+0x0/0x118) from [<c0577df8>] (rest_init+0x8c/0xa4) [ 3354.616707] r7:c130e1c0 r6:c077abe4 r5:c0817880 r4:c0788000 [ 3354.616735] [<c0577d6c>] (rest_init+0x0/0xa4) from [<c074e82c>] (start_kernel+0x2d0/0x324) [ 3354.616744] r4:c07aa0ac r3:00000000 [ 3354.616761] [<c074e55c>] (start_kernel+0x0/0x324) from [<80008040>] (0x80008040) [ 3354.616769] r7:c07adf44 r6:c077abe0 r5:c07a9650 r4:10c5387d [ 3354.616785] ---[ end trace 07edee490892b78c ]--- [ 3354.616797] e1000e 0000:01:00.0: eth0: NETDEV WATCHDOG: transmit timed out - resetting -------------------------------------------------------------------- On 18/06/14 17:00, tony wrote: > Hi, > I am getting link drops in my tegra based board with 82574l > NIC. The OS is Linux Ubuntu 13.04. The e1000e driver was upgraded to > 3.04 to rule out any software problems. I am using iperf to reproduce > the problem. > > The 2 platforms which runs iperf server and client are identical boards > with 82574l. I assign them static IP connected to each other using CAT5e > cable. From various forums, I read there are problems with ASPM with the > NIC. I have disabled ASPM in the kernel as well (just to be safe, even > if it already disabled in the driver). I have also tried disabling TSO. > But still can reproduce the problem. > > Attached are the logs (TSO not disabled). > > I couldn't find any error messages from the kernel nor the driver other > than the watchdog resetting the driver. Any pointers on how to fix the > link will be very much appreciated. > > Thanks, > Tony > > > ------------------------------------------------------------------------------ > HPCC Systems Open Source Big Data Platform from LexisNexis Risk Solutions > Find What Matters Most in Your Big Data with HPCC Systems > Open Source. Fast. Scalable. Simple. Ideal for Dirty Data. > Leverages Graph Analysis for Fast Processing & Easy Data Exploration > http://p.sf.net/sfu/hpccsystems > > > > _______________________________________________ > E1000-devel mailing list > E1000-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/e1000-devel > To learn more about Intel® Ethernet, visit > http://communities.intel.com/community/wired > ------------------------------------------------------------------------------ HPCC Systems Open Source Big Data Platform from LexisNexis Risk Solutions Find What Matters Most in Your Big Data with HPCC Systems Open Source. Fast. Scalable. Simple. Ideal for Dirty Data. Leverages Graph Analysis for Fast Processing & Easy Data Exploration http://p.sf.net/sfu/hpccsystems _______________________________________________ E1000-devel mailing list E1000-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/e1000-devel To learn more about Intel® Ethernet, visit http://communities.intel.com/community/wired