Hello everyone, We've been experiencing a slew of issues with our 82574L chips and the e1000e driver. We have several of these units in the field. We're using a vanilla kernel 2.6.27.49 with e1000e 2.1.4. Kernel and hardware supports MSI-X but changing these options (IntMode=1 and disabling MSI-X globally) has no effect. Kernel is booted with pcie_aspm=off and we run the following setpci commands on boot:
setpci -s 01:00.0 CAP_EXP+10.b=40 setpci -s 02:00.0 CAP_EXP+10.b=40 setpci -s 03:00.0 CAP_EXP+10.b=40 setpci -s 04:00.0 CAP_EXP+10.b=40 We (typically) discover the eeprom corruption issue addressed with this patch as well: http://sourceforge.net/projects/e1000/files/e1000e%20stable/eeprom_fix_82574_or_82583/ and apply it to affected units (even though our kernel isn't compiled with FTRACE): grep -i ftrace .config CONFIG_HAVE_FTRACE=y CONFIG_HAVE_DYNAMIC_FTRACE=y # CONFIG_FTRACE is not set e1000e driver is compiled with "EXTRA_CFLAGS=-DDEBUG" . Even with all of these fixes and precautions we find that we typically "lose" at least one (if not two) out of the four interfaces on this system: e1000e 0000:01:00.0: eth0: Masking off all interrupts e1000e 0000:01:00.0: eth0: Issuing a global reset to MAC e1000e 0000:01:00.0: eth0: Initializing the IEEE VLAN e1000e 0000:01:00.0: eth0: Programming MAC Address into RAR[0] e1000e 0000:01:00.0: eth0: Clearing RAR[1-14] e1000e 0000:01:00.0: eth0: Zeroing the MTA e1000e 0000:01:00.0: eth0: After fix-ups FlowControl is now = 3 e1000e 0000:01:00.0: eth0: Reconfiguring auto-neg advertisement params e1000e 0000:01:00.0: eth0: autoneg_advertised 2f e1000e 0000:01:00.0: eth0: Advertise 10mb Half duplex e1000e 0000:01:00.0: eth0: Advertise 10mb Full duplex e1000e 0000:01:00.0: eth0: Advertise 100mb Half duplex e1000e 0000:01:00.0: eth0: Advertise 100mb Full duplex e1000e 0000:01:00.0: eth0: Advertise 1000mb Full duplex e1000e 0000:01:00.0: eth0: Auto-Neg Advertising de1 e1000e 0000:01:00.0: eth0: Restarting Auto-Neg e1000e 0000:01:00.0: eth0: Unable to establish link!!! e1000e 0000:01:00.0: eth0: Initializing the Flow Control address, type and timer regs e1000e 0000:01:00.0: PCI INT A disabled e1000e 0000:01:00.0: Disabling ASPM L0s L1 e1000e 0000:01:00.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16 e1000e 0000:01:00.0: setting latency timer to 64 e1000e 0000:01:00.0: Interrupt Throttling Rate (ints/sec) set to dynamic conservative mode e1000e 0000:01:00.0: (unregistered net_device): Masking off all interrupts e1000e 0000:01:00.0: (unregistered net_device): Issuing a global reset to MAC e1000e 0000:01:00.0: (unregistered net_device): Masking off all interrupts e1000e 0000:01:00.0: (unregistered net_device): Issuing a global reset to MAC e1000e 0000:01:00.0: (unregistered net_device): Initializing the IEEE VLAN e1000e 0000:01:00.0: (unregistered net_device): Programming MAC Address into RAR[0] e1000e 0000:01:00.0: (unregistered net_device): Clearing RAR[1-14] e1000e 0000:01:00.0: (unregistered net_device): Zeroing the MTA e1000e 0000:01:00.0: (unregistered net_device): After fix-ups FlowControl is now = 3 e1000e 0000:01:00.0: (unregistered net_device): Reconfiguring auto-neg advertisement params e1000e 0000:01:00.0: (unregistered net_device): autoneg_advertised 2f e1000e 0000:01:00.0: (unregistered net_device): Advertise 10mb Half duplex e1000e 0000:01:00.0: (unregistered net_device): Advertise 10mb Full duplex e1000e 0000:01:00.0: (unregistered net_device): Advertise 100mb Half duplex e1000e 0000:01:00.0: (unregistered net_device): Advertise 100mb Full duplex e1000e 0000:01:00.0: (unregistered net_device): Advertise 1000mb Full duplex e1000e 0000:01:00.0: (unregistered net_device): Auto-Neg Advertising de1 e1000e 0000:01:00.0: (unregistered net_device): Restarting Auto-Neg e1000e 0000:01:00.0: (unregistered net_device): Unable to establish link!!! e1000e 0000:01:00.0: (unregistered net_device): Initializing the Flow Control address, type and timer regs e1000e 0000:01:00.0: eth0: (PCI Express:2.5GT/s:Width x1) 4c:02:89:08:ca:02 e1000e 0000:01:00.0: eth0: Intel(R) PRO/1000 Network Connection e1000e 0000:01:00.0: eth0: NVM PBA number is not stored as string e1000e 0000:01:00.0: eth0: MAC: 4, PHY: 8, PBA No: FFFFFF-0FF e1000e 0000:02:00.0: Disabling ASPM L0s L1 e1000e 0000:02:00.0: PCI INT A -> GSI 17 (level, low) -> IRQ 17 e1000e 0000:02:00.0: setting latency timer to 64 e1000e 0000:02:00.0: Interrupt Throttling Rate (ints/sec) set to dynamic conservative mode e1000e 0000:02:00.0: (unregistered net_device): Masking off all interrupts e1000e 0000:02:00.0: (unregistered net_device): Issuing a global reset to MAC e1000e 0000:02:00.0: (unregistered net_device): Masking off all interrupts e1000e 0000:02:00.0: (unregistered net_device): Issuing a global reset to MAC e1000e 0000:02:00.0: (unregistered net_device): Initializing the IEEE VLAN e1000e 0000:02:00.0: (unregistered net_device): Programming MAC Address into RAR[0] e1000e 0000:02:00.0: (unregistered net_device): Clearing RAR[1-14] e1000e 0000:02:00.0: (unregistered net_device): Zeroing the MTA e1000e 0000:02:00.0: (unregistered net_device): After fix-ups FlowControl is now = 3 e1000e 0000:02:00.0: (unregistered net_device): Reconfiguring auto-neg advertisement params e1000e 0000:02:00.0: (unregistered net_device): autoneg_advertised 2f e1000e 0000:02:00.0: (unregistered net_device): Advertise 10mb Half duplex e1000e 0000:02:00.0: (unregistered net_device): Advertise 10mb Full duplex e1000e 0000:02:00.0: (unregistered net_device): Advertise 100mb Half duplex e1000e 0000:02:00.0: (unregistered net_device): Advertise 100mb Full duplex e1000e 0000:02:00.0: (unregistered net_device): Advertise 1000mb Full duplex e1000e 0000:02:00.0: (unregistered net_device): Auto-Neg Advertising de1 e1000e 0000:02:00.0: (unregistered net_device): Restarting Auto-Neg e1000e 0000:02:00.0: (unregistered net_device): Unable to establish link!!! e1000e 0000:02:00.0: (unregistered net_device): Initializing the Flow Control address, type and timer regs e1000e 0000:02:00.0: eth1: (PCI Express:2.5GT/s:Width x1) 4c:02:89:08:ca:03 e1000e 0000:02:00.0: eth1: Intel(R) PRO/1000 Network Connection e1000e 0000:02:00.0: eth1: NVM PBA number is not stored as string e1000e 0000:02:00.0: eth1: MAC: 4, PHY: 8, PBA No: FFFFFF-0FF e1000e 0000:03:00.0: Disabling ASPM L0s L1 e1000e 0000:03:00.0: PCI INT A -> GSI 18 (level, low) -> IRQ 18 e1000e 0000:03:00.0: setting latency timer to 64 e1000e 0000:03:00.0: Interrupt Throttling Rate (ints/sec) set to dynamic conservative mode e1000e 0000:03:00.0: (unregistered net_device): Driver can't access the PHY e1000e 0000:03:00.0: (unregistered net_device): Error getting PHY ID e1000e 0000:03:00.0: PCI INT A disabled e1000e: probe of 0000:03:00.0 failed with error -2 e1000e 0000:04:00.0: Disabling ASPM L0s L1 e1000e 0000:04:00.0: PCI INT A -> GSI 19 (level, low) -> IRQ 19 e1000e 0000:04:00.0: setting latency timer to 64 e1000e 0000:04:00.0: Interrupt Throttling Rate (ints/sec) set to dynamic conservative mode e1000e 0000:04:00.0: (unregistered net_device): Driver can't access the PHY e1000e 0000:04:00.0: (unregistered net_device): Error getting PHY ID e1000e 0000:04:00.0: PCI INT A disabled e1000e: probe of 0000:04:00.0 failed with error -2 "Driver can't access the PHY" and "Error getting PHY ID". At this point it is virtually impossible to recover the unit and we have to recall it. What can I do to address this? Thanks! -- Kristian Kielhofner ------------------------------------------------------------------------------ Everyone hates slow websites. So do we. Make your web apps faster with AppDynamics Download AppDynamics Lite for free today: http://ad.doubleclick.net/clk;258768047;13503038;j? http://info.appdynamics.com/FreeJavaPerformanceDownload.html _______________________________________________ E1000-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/e1000-devel To learn more about Intel® Ethernet, visit http://communities.intel.com/community/wired
