[Kernel-packages] [Bug 1898057] Re: Infiniband transmit queue timeouts after upgrading to linux-hwe-5.4
apport information ** Package changed: linux-hwe-5.4 (Ubuntu) => linux (Ubuntu) ** Tags added: apport-collected bionic ** Description changed: After upgrading 3 servers from linux-image-5.3.0-40-generic to linux- image-5.4.0-48-generic I have started seeing the following queue timeouts from IP-over-Infiniband (ipoib) devices. The devices in question are (with newest available firmware, 2.42.5000): # lspci -nnk -s 83:00.0 83:00.0 Network controller [0280]: Mellanox Technologies MT27500 Family [ConnectX-3] [15b3:1003] Subsystem: Mellanox Technologies MT27500 Family [ConnectX-3] [15b3:0027] Kernel driver in use: mlx4_core Below is the WARN from one machine's syslog. The others are practically identical. When the WARN happens on any of the machines, other 2 will _also_ exhibit queue timeouts. Additionally, other (unrelated) machines connected to the same infiniband fabric will exhibit a 12-second transmission delay. This could conceivably be caused by these 3 servers also being Subnet Managers (opensm package). The infiniband fabric is partitioned, with the affected partition (8011) seeing most of the traffic. kernel: [52642.480066] [ cut here ] kernel: [52642.480092] NETDEV WATCHDOG: ib0.8011 (): transmit queue 0 timed out kernel: [52642.480120] WARNING: CPU: 13 PID: 0 at /build/linux-hwe-5.4-8m2I8l/linux-hwe-5.4-5.4.0/net/sched/sch_generic.c:448 dev_watchdog+0x264/0x270 kernel: [52642.480121] Modules linked in: aufs overlay ip6table_raw ip6table_mangle ip6table_nat iptable_raw iptable_mangle iptable_nat nf_tables nfnetlink cfg80211 ebtable_filter ebtables ip6table_filter ip6_tables iptable_filter bpfilter mst_pciconf(OE) mst_pci(OE) 8021q garp mrp stp llc nls_iso8859_1 intel_rapl_msr lz4 lz4_compress intel_rapl_common ib_iser rdma_cm sb_edac iw_cm iscsi_tcp libiscsi_tcp libiscsi x86_pkg_temp_thermal scsi_transport_iscsi intel_powerclamp zram veth vhost_net tap coretemp vhost kvm_intel crct10dif_pclmul crc32_pclmul ghash_clmulni_intel kvm openvswitch nsh nf_conncount nf_nat nf_conntrack rapl nf_defrag_ipv6 nf_defrag_ipv4 joydev input_leds intel_cstate ib_ipoib mei_me mei ib_cm ioatdma ib_umad lpc_ich acpi_pad acpi_power_meter mac_hid ipmi_si ipmi_ssif ipmi_devintf ipmi_msghandler kyber_iosched sch_fq_codel tcp_highspeed ip_tables x_tables autofs4 zfs(POE) zunicode(POE) zavl(POE) icp(POE) zcommon(POE) znvpair(POE) spl(OE) zlua(POE) btrfs zstd_compress raid10 raid456 kernel: [52642.480182] async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid0 multipath linear dm_mirror dm_region_hash dm_log mlx4_ib ib_uverbs ib_core hid_generic raid1 ses enclosure usbhid hid ast drm_vram_helper i2c_algo_bit ttm drm_kms_helper syscopyarea sysfillrect sysimgblt aesni_intel fb_sys_fops ixgbe glue_helper mpt3sas nvme xfrm_algo crypto_simd ahci raid_class dca cryptd mlx4_core drm megaraid_sas nvme_core libahci scsi_transport_sas mdio wmi kernel: [52642.480221] CPU: 13 PID: 0 Comm: swapper/13 Tainted: P OE 5.4.0-48-generic #52~18.04.1-Ubuntu kernel: [52642.480223] Hardware name: Supermicro Super Server/X10DRW-iT, BIOS 2.0b 04/13/2017 kernel: [52642.480226] RIP: 0010:dev_watchdog+0x264/0x270 kernel: [52642.480229] Code: 48 85 c0 75 e6 eb a0 4c 89 ef c6 05 42 c1 e7 00 01 e8 30 b8 fa ff 89 d9 48 89 c2 4c 89 ee 48 c7 c7 50 05 63 ae e8 4c 31 71 ff <0f> 0b eb 82 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 55 48 89 e5 41 kernel: [52642.480230] RSP: 0018:b4998c970e48 EFLAGS: 00010282 kernel: [52642.480233] RAX: RBX: RCX: 083f kernel: [52642.480234] RDX: RSI: 00f6 RDI: 083f kernel: [52642.480235] RBP: b4998c970e78 R08: 08fd R09: 0003 kernel: [52642.480237] R10: b4998c970ee8 R11: 0001 R12: 0001 kernel: [52642.480238] R13: 93d302465000 R14: 93d302465480 R15: 93f2d293c880 kernel: [52642.480240] FS: () GS:93f33f64() knlGS: kernel: [52642.480242] CS: 0010 DS: ES: CR0: 80050033 kernel: [52642.480243] CR2: f90140127000 CR3: 002a26e0a004 CR4: 001626e0 kernel: [52642.480245] Call Trace: kernel: [52642.480247] kernel: [52642.480252] ? pfifo_fast_reset+0x110/0x110 kernel: [52642.480255] call_timer_fn+0x32/0x130 kernel: [52642.480258] run_timer_softirq+0x443/0x480 kernel: [52642.480262] ? ktime_get+0x43/0xa0 kernel: [52642.480268] ? lapic_next_deadline+0x26/0x30 kernel: [52642.480273] __do_softirq+0xe4/0x2da kernel: [52642.480278] irq_exit+0xae/0xb0 kernel: [52642.480282] smp_apic_timer_interrupt+0x79/0x130 kernel: [52642.480285] apic_timer_interrupt+0xf/0x20 kernel: [52642.480286] kernel: [52642.480292] RIP: 0010:cpuidle_enter_state+0xbc/0x440 kernel
[Kernel-packages] [Bug 1898057] Re: Infiniband transmit queue timeouts after upgrading to linux-hwe-5.4
** Changed in: linux (Ubuntu) Status: New => Confirmed -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux in Ubuntu. https://bugs.launchpad.net/bugs/1898057 Title: Infiniband transmit queue timeouts after upgrading to linux-hwe-5.4 Status in linux package in Ubuntu: Confirmed Bug description: After upgrading 3 servers from linux-image-5.3.0-40-generic to linux- image-5.4.0-48-generic I have started seeing the following queue timeouts from IP-over-Infiniband (ipoib) devices. The devices in question are (with newest available firmware, 2.42.5000): # lspci -nnk -s 83:00.0 83:00.0 Network controller [0280]: Mellanox Technologies MT27500 Family [ConnectX-3] [15b3:1003] Subsystem: Mellanox Technologies MT27500 Family [ConnectX-3] [15b3:0027] Kernel driver in use: mlx4_core Below is the WARN from one machine's syslog. The others are practically identical. When the WARN happens on any of the machines, other 2 will _also_ exhibit queue timeouts. Additionally, other (unrelated) machines connected to the same infiniband fabric will exhibit a 12-second transmission delay. This could conceivably be caused by these 3 servers also being Subnet Managers (opensm package). The infiniband fabric is partitioned, with the affected partition (8011) seeing most of the traffic. kernel: [52642.480066] [ cut here ] kernel: [52642.480092] NETDEV WATCHDOG: ib0.8011 (): transmit queue 0 timed out kernel: [52642.480120] WARNING: CPU: 13 PID: 0 at /build/linux-hwe-5.4-8m2I8l/linux-hwe-5.4-5.4.0/net/sched/sch_generic.c:448 dev_watchdog+0x264/0x270 kernel: [52642.480121] Modules linked in: aufs overlay ip6table_raw ip6table_mangle ip6table_nat iptable_raw iptable_mangle iptable_nat nf_tables nfnetlink cfg80211 ebtable_filter ebtables ip6table_filter ip6_tables iptable_filter bpfilter mst_pciconf(OE) mst_pci(OE) 8021q garp mrp stp llc nls_iso8859_1 intel_rapl_msr lz4 lz4_compress intel_rapl_common ib_iser rdma_cm sb_edac iw_cm iscsi_tcp libiscsi_tcp libiscsi x86_pkg_temp_thermal scsi_transport_iscsi intel_powerclamp zram veth vhost_net tap coretemp vhost kvm_intel crct10dif_pclmul crc32_pclmul ghash_clmulni_intel kvm openvswitch nsh nf_conncount nf_nat nf_conntrack rapl nf_defrag_ipv6 nf_defrag_ipv4 joydev input_leds intel_cstate ib_ipoib mei_me mei ib_cm ioatdma ib_umad lpc_ich acpi_pad acpi_power_meter mac_hid ipmi_si ipmi_ssif ipmi_devintf ipmi_msghandler kyber_iosched sch_fq_codel tcp_highspeed ip_tables x_tables autofs4 zfs(POE) zunicode(POE) zavl(POE) icp(POE) zcommon(POE) znvpair(POE) spl(OE) zlua(POE) btrfs zstd_compress raid10 raid456 kernel: [52642.480182] async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid0 multipath linear dm_mirror dm_region_hash dm_log mlx4_ib ib_uverbs ib_core hid_generic raid1 ses enclosure usbhid hid ast drm_vram_helper i2c_algo_bit ttm drm_kms_helper syscopyarea sysfillrect sysimgblt aesni_intel fb_sys_fops ixgbe glue_helper mpt3sas nvme xfrm_algo crypto_simd ahci raid_class dca cryptd mlx4_core drm megaraid_sas nvme_core libahci scsi_transport_sas mdio wmi kernel: [52642.480221] CPU: 13 PID: 0 Comm: swapper/13 Tainted: P OE 5.4.0-48-generic #52~18.04.1-Ubuntu kernel: [52642.480223] Hardware name: Supermicro Super Server/X10DRW-iT, BIOS 2.0b 04/13/2017 kernel: [52642.480226] RIP: 0010:dev_watchdog+0x264/0x270 kernel: [52642.480229] Code: 48 85 c0 75 e6 eb a0 4c 89 ef c6 05 42 c1 e7 00 01 e8 30 b8 fa ff 89 d9 48 89 c2 4c 89 ee 48 c7 c7 50 05 63 ae e8 4c 31 71 ff <0f> 0b eb 82 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 55 48 89 e5 41 kernel: [52642.480230] RSP: 0018:b4998c970e48 EFLAGS: 00010282 kernel: [52642.480233] RAX: RBX: RCX: 083f kernel: [52642.480234] RDX: RSI: 00f6 RDI: 083f kernel: [52642.480235] RBP: b4998c970e78 R08: 08fd R09: 0003 kernel: [52642.480237] R10: b4998c970ee8 R11: 0001 R12: 0001 kernel: [52642.480238] R13: 93d302465000 R14: 93d302465480 R15: 93f2d293c880 kernel: [52642.480240] FS: () GS:93f33f64() knlGS: kernel: [52642.480242] CS: 0010 DS: ES: CR0: 80050033 kernel: [52642.480243] CR2: f90140127000 CR3: 002a26e0a004 CR4: 001626e0 kernel: [52642.480245] Call Trace: kernel: [52642.480247] kernel: [52642.480252] ? pfifo_fast_reset+0x110/0x110 kernel: [52642.480255] call_timer_fn+0x32/0x130 kernel: [52642.480258] run_timer_softirq+0x443/0x480 kernel: [52642.480262] ? ktime_get+0x43/0xa0 kernel: [52642.480268] ? lapic_next_deadline+0x26/0x30 kernel: [52642.480273] __do_softirq+0xe4/0x2da kernel: [52642.480278] irq_exit+0xae/0xb0 ker
[Kernel-packages] [Bug 1898057] Re: Infiniband transmit queue timeouts after upgrading to linux-hwe-5.4
After a reboot of the whole fabric (switches and machines), this problem has not resurfaced. Conceivably this could have been a transient error state, so I will close this with "invalid". ** Changed in: linux (Ubuntu) Status: Confirmed => Invalid -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux in Ubuntu. https://bugs.launchpad.net/bugs/1898057 Title: Infiniband transmit queue timeouts after upgrading to linux-hwe-5.4 Status in linux package in Ubuntu: Invalid Bug description: After upgrading 3 servers from linux-image-5.3.0-40-generic to linux- image-5.4.0-48-generic I have started seeing the following queue timeouts from IP-over-Infiniband (ipoib) devices. The devices in question are (with newest available firmware, 2.42.5000): # lspci -nnk -s 83:00.0 83:00.0 Network controller [0280]: Mellanox Technologies MT27500 Family [ConnectX-3] [15b3:1003] Subsystem: Mellanox Technologies MT27500 Family [ConnectX-3] [15b3:0027] Kernel driver in use: mlx4_core Below is the WARN from one machine's syslog. The others are practically identical. When the WARN happens on any of the machines, other 2 will _also_ exhibit queue timeouts. Additionally, other (unrelated) machines connected to the same infiniband fabric will exhibit a 12-second transmission delay. This could conceivably be caused by these 3 servers also being Subnet Managers (opensm package). The infiniband fabric is partitioned, with the affected partition (8011) seeing most of the traffic. kernel: [52642.480066] [ cut here ] kernel: [52642.480092] NETDEV WATCHDOG: ib0.8011 (): transmit queue 0 timed out kernel: [52642.480120] WARNING: CPU: 13 PID: 0 at /build/linux-hwe-5.4-8m2I8l/linux-hwe-5.4-5.4.0/net/sched/sch_generic.c:448 dev_watchdog+0x264/0x270 kernel: [52642.480121] Modules linked in: aufs overlay ip6table_raw ip6table_mangle ip6table_nat iptable_raw iptable_mangle iptable_nat nf_tables nfnetlink cfg80211 ebtable_filter ebtables ip6table_filter ip6_tables iptable_filter bpfilter mst_pciconf(OE) mst_pci(OE) 8021q garp mrp stp llc nls_iso8859_1 intel_rapl_msr lz4 lz4_compress intel_rapl_common ib_iser rdma_cm sb_edac iw_cm iscsi_tcp libiscsi_tcp libiscsi x86_pkg_temp_thermal scsi_transport_iscsi intel_powerclamp zram veth vhost_net tap coretemp vhost kvm_intel crct10dif_pclmul crc32_pclmul ghash_clmulni_intel kvm openvswitch nsh nf_conncount nf_nat nf_conntrack rapl nf_defrag_ipv6 nf_defrag_ipv4 joydev input_leds intel_cstate ib_ipoib mei_me mei ib_cm ioatdma ib_umad lpc_ich acpi_pad acpi_power_meter mac_hid ipmi_si ipmi_ssif ipmi_devintf ipmi_msghandler kyber_iosched sch_fq_codel tcp_highspeed ip_tables x_tables autofs4 zfs(POE) zunicode(POE) zavl(POE) icp(POE) zcommon(POE) znvpair(POE) spl(OE) zlua(POE) btrfs zstd_compress raid10 raid456 kernel: [52642.480182] async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid0 multipath linear dm_mirror dm_region_hash dm_log mlx4_ib ib_uverbs ib_core hid_generic raid1 ses enclosure usbhid hid ast drm_vram_helper i2c_algo_bit ttm drm_kms_helper syscopyarea sysfillrect sysimgblt aesni_intel fb_sys_fops ixgbe glue_helper mpt3sas nvme xfrm_algo crypto_simd ahci raid_class dca cryptd mlx4_core drm megaraid_sas nvme_core libahci scsi_transport_sas mdio wmi kernel: [52642.480221] CPU: 13 PID: 0 Comm: swapper/13 Tainted: P OE 5.4.0-48-generic #52~18.04.1-Ubuntu kernel: [52642.480223] Hardware name: Supermicro Super Server/X10DRW-iT, BIOS 2.0b 04/13/2017 kernel: [52642.480226] RIP: 0010:dev_watchdog+0x264/0x270 kernel: [52642.480229] Code: 48 85 c0 75 e6 eb a0 4c 89 ef c6 05 42 c1 e7 00 01 e8 30 b8 fa ff 89 d9 48 89 c2 4c 89 ee 48 c7 c7 50 05 63 ae e8 4c 31 71 ff <0f> 0b eb 82 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 55 48 89 e5 41 kernel: [52642.480230] RSP: 0018:b4998c970e48 EFLAGS: 00010282 kernel: [52642.480233] RAX: RBX: RCX: 083f kernel: [52642.480234] RDX: RSI: 00f6 RDI: 083f kernel: [52642.480235] RBP: b4998c970e78 R08: 08fd R09: 0003 kernel: [52642.480237] R10: b4998c970ee8 R11: 0001 R12: 0001 kernel: [52642.480238] R13: 93d302465000 R14: 93d302465480 R15: 93f2d293c880 kernel: [52642.480240] FS: () GS:93f33f64() knlGS: kernel: [52642.480242] CS: 0010 DS: ES: CR0: 80050033 kernel: [52642.480243] CR2: f90140127000 CR3: 002a26e0a004 CR4: 001626e0 kernel: [52642.480245] Call Trace: kernel: [52642.480247] kernel: [52642.480252] ? pfifo_fast_reset+0x110/0x110 kernel: [52642.480255] call_timer_fn+0x32/0x130 kernel: [52642.480258] run_timer_softirq+0x443/0x480 kernel: [5264