2012/9/13 Dave, Tushar N <tushar.n.d...@intel.com>: >>-----Original Message----- >>From: hank peng [mailto:pengxi...@gmail.com] >>Sent: Tuesday, September 11, 2012 5:40 PM >>To: e1000-devel >>Subject: [E1000-devel] 82574L crashed in heavy load test >> >>Kernel version is 2.6.37.1, network driver version is 2.0.0.1. We test its >>stability in heavy load test. After about several minutes, it crashed. >>Here is messages: >>[ 1231.082585] ------------[ cut here ]------------ [ 1231.087211] >>WARNING: at net/sched/sch_generic.c:258 dev_watchdog+0xdc/0x133() [ >>1231.094336] Hardware name: SugarBay [ 1231.097820] NETDEV WATCHDOG: eth0 >>(e1000e): transmit queue 0 timed out [ 1231.104335] Modules linked in: >>iscsi_trgt e1000e [last unloaded: iscsi_trgt] [ 1231.111440] Pid: 0, comm: >>swapper Not tainted 2.6.37.1 #8 [ 1231.116829] Call Trace: >>[ 1231.119275] [<c1032190>] warn_slowpath_common+0x60/0x75 [ 1231.124586] >>[<c1032209>] warn_slowpath_fmt+0x26/0x2a [ 1231.129627] [<c13eaef2>] >>dev_watchdog+0xdc/0x133 [ 1231.134325] [<c102be88>] ? >>rebalance_domains+0xb3/0x112 [ 1231.139629] [<c103bc04>] >>run_timer_softirq+0x12c/0x1cf [ 1231.144853] [<c13eae16>] ? >>dev_watchdog+0x0/0x133 [ 1231.149645] [<c1036c4e>] >>__do_softirq+0xab/0x14c [ 1231.154348] [<c1036ba3>] ? >>__do_softirq+0x0/0x14c [ 1231.159131] <IRQ> [<c1036abf>] ? >>irq_exit+0x2d/0x61 [ 1231.164192] [<c1017ad8>] ? >>smp_apic_timer_interrupt+0x60/0x70 >>[ 1231.170015] [<c14a80aa>] ? apic_timer_interrupt+0x2a/0x30 [ >>1231.175493] [<c129ab20>] ? acpi_idle_enter_bm+0x20c/0x244 [ 1231.180976] >>[<c13bab21>] ? cpuidle_idle_call+0x71/0xd0 [ 1231.186201] [<c1001c94>] ? >>cpu_idle+0x45/0x91 [ 1231.190646] [<c149655c>] ? rest_init+0x58/0x5a [ >>1231.195177] [<c16c98e2>] ? start_kernel+0x2c5/0x2ca [ 1231.200134] >>[<c16c90b8>] ? i386_start_kernel+0xb8/0xbd [ 1231.205358] ---[ end trace >>e04d69be6eefa026 ]--- [ 1231.210182] e1000e 0000:08:00.0: eth0: Reset >>adapter > > Please send me the full dmesg log after issue occurs. > How often does the issue occurs? > What kind of traffic are you running in your heavy load test? ( Are you > running short packet test?) > Send output of ethtool -S ethx after issue occurs. > Send lspci -vvv output after issue occurs. > It is not easy to reproduce this issue. We are testing the latest driver, and I noticed that the latest driver has disabled 82574's L1 and L0s ASPM. In addition, we disabled the CONFIG_PCIE_ASPM in kernel's config.
Now, lspci -vvv shows that 81574's L1 and L0s ASPM are all disabled. We are continuing to test, I will report if this issue occured. > -Tushar > > -- The simplest is not all best but the best is surely the simplest! ------------------------------------------------------------------------------ Got visibility? Most devs has no idea what their production app looks like. Find out how fast your code is with AppDynamics Lite. http://ad.doubleclick.net/clk;262219671;13503038;y? http://info.appdynamics.com/FreeJavaPerformanceDownload.html _______________________________________________ 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
