On Fri, 2018-07-27 at 08:05 -0400, Laurence Oberman wrote: > On Thu, 2018-07-26 at 16:02 -0400, Laurence Oberman wrote: > > On Thu, 2018-07-26 at 10:28 -0400, Don Dutile wrote: > > > On 07/26/2018 08:48 AM, Laurence Oberman wrote: > > > > Hello > > > > > > > > https://www.spinics.net/lists/linux-rdma/msg51334.html > > > > > > > > A rhel 7.5 with backports from upstream is hitting this. > > > > Chuck Reported it and Sagi and Max responded but its not clear > > > > if > > > > we > > > > ever fixed this. > > > > > > > > > > RHEL-7.5 data point: > > > -- drivers/infiniband/* -r is backported to v4.14. > > > i.e., includes the patch(es) mentioned in the above thread. > > > > > > Laurence: > > > Please test with 7.6 kernel & report back. > > > if that passes, RH can bisect the bug fix btwn v4.14 & v4.16(the > > > 7.6 > > > update point for its rdma kernel core), > > > and backport to 7.5-zstream. note: you'll have to update rdma- > > > core > > > pkg to the 7.6 version as well. > > > All functional & bug fix patches to mlx* (ib & enet) are in as > > > well > > > (same kernel references). > > > > > > -dd > > > > > > > In this case we land up in a panic, noty just messaging, > > > > although > > > > the > > > > messages logged for a long time over and over until we finally > > > > panicked. > > > > > > > > crash> log | grep "memreg failure: memor" | wc -l > > > > 2414 > > > > > > > > crash> log > > > > [1635578.012721] connection16:0: detected conn error (1011) > > > > [1635587.050688] mlx5_0:dump_cqe:262:(pid 93128): dump error > > > > cqe > > > > [1635587.089686] 00000000 00000000 00000000 00000000 > > > > [1635587.123989] 00000000 00000000 00000000 00000000 > > > > [1635587.157494] 00000000 00000000 00000000 00000000 > > > > [1635587.190968] 00000000 08007806 250002ad ba6115d3 > > > > > > > > [1635587.224331] iser: iser_err_comp: memreg failure: memory > > > > management > > > > operation error (6) vend_err 78 > > > > [1635587.278876] connection15:0: detected conn error (1011) > > > > [1635590.986286] mlx5_1:dump_cqe:262:(pid 0): dump error cqe > > > > [1635591.021891] 00000000 00000000 00000000 00000000 > > > > [1635591.053944] 00000000 00000000 00000000 00000000 > > > > > > > > [1657077.997960] BUG: unable to handle kernel NULL pointer > > > > dereference > > > > at 0000000000000010 > > > > [1657077.997967] IP: [<ffffffffc08a541e>] > > > > iscsi_verify_itt+0x1e/0x110 > > > > [libiscsi] > > > > [1657077.997970] PGD 80000098de387067 PUD b8d9ffa067 PMD 0 > > > > [1657077.997971] Oops: 0000 [#1] SMP > > > > [1657077.998009] Modules linked in: oracleasm(O) nfsv3 > > > > rpcsec_gss_krb5 > > > > nfsv4 dns_resolver nfs fscache dm_round_robin bonding rpcrdma > > > > ib_isert > > > > iscsi_target_mod ib_iser libiscsi scsi_transport_iscsi ib_srpt > > > > target_core_mod ib_srp scsi_transport_srp scsi_tgt ib_ipoib > > > > rdma_ucm > > > > ib_ucm ib_uverbs ib_umad rdma_cm ib_cm iw_cm mlx5_ib ib_core > > > > vfat > > > > fat > > > > xfs sb_edac edac_core intel_powerclamp coretemp intel_rapl > > > > iosf_mbi > > > > kvm_intel kvm irqbypass iTCO_wdt crc32_pclmul ipmi_ssif > > > > iTCO_vendor_support ghash_clmulni_intel aesni_intel lrw > > > > gf128mul > > > > ipmi_si glue_helper ablk_helper cryptd sg hpwdt hpilo pcspkr > > > > ipmi_devintf ioatdma dm_multipath i2c_i801 lpc_ich shpchp dca > > > > wmi > > > > ipmi_msghandler pcc_cpufreq acpi_power_meter nfsd binfmt_misc > > > > auth_rpcgss nfs_acl lockd grace sunrpc ip_tables ext4 mbcache > > > > jbd2 > > > > sd_mod crc_t10dif crct10dif_generic > > > > [1657077.998020] i2c_algo_bit drm_kms_helper syscopyarea > > > > sysfillrect > > > > sysimgblt fb_sys_fops ttm bnx2x mlx5_core crct10dif_pclmul mdio > > > > tg3(OE) > > > > devlink libcrc32c crct10dif_common drm hpsa(OE) ptp i2c_core > > > > crc32c_intel scsi_transport_sas pps_core dm_mirror > > > > dm_region_hash > > > > dm_log dm_mod > > > > [1657077.998023] CPU: 20 PID: 41538 Comm: sh Tainted: > > > > G OE - > > > > ----------- 3.10.0-693.34.1.el7_bz1582551.x86_64 #1 > > > > [1657077.998024] Hardware name: HP ProLiant DL380 Gen9/ProLiant > > > > DL380 > > > > Gen9, BIOS P89 05/21/2018 > > > > [1657077.998025] task: ffff88587ce38fd0 ti: ffff884dd0af0000 > > > > task.ti: > > > > ffff884dd0af0000 > > > > [1657077.998029] RIP: > > > > 0010:[<ffffffffc08a541e>] [<ffffffffc08a541e>] > > > > iscsi_verify_itt+0x1e/0x110 [libiscsi] > > > > [1657077.998030] RSP: 0000:ffff88beff403d78 EFLAGS: 00010286 > > > > [1657077.998031] RAX: 000000000000004c RBX: 00000000b0000036 > > > > RCX: > > > > 0000000000000002 > > > > [1657077.998032] RDX: 00000000000000cc RSI: 00000000b0000036 > > > > RDI: > > > > 0000000000000000 > > > > [1657077.998033] RBP: ffff88beff403da0 R08: 0000000040032a20 > > > > R09: > > > > ffff8896e4eaf91c > > > > [1657077.998034] R10: 0000000000000000 R11: 00007ffff7763ca0 > > > > R12: > > > > 0000000000000000 > > > > [1657077.998035] R13: ffff8896e4eaf9e4 R14: ffff8896e4eaf900 > > > > R15: > > > > 0000000000000000 > > > > [1657077.998036] FS: 00007ffff7fe6740(0000) > > > > GS:ffff88beff400000(0000) > > > > knlGS:0000000000000000 > > > > [1657077.998038] CS: 0010 DS: 0000 ES: 0000 CR0: > > > > 0000000080050033 > > > > [1657077.998039] CR2: 0000000000000010 CR3: 000000ad92eba000 > > > > CR4: > > > > 00000000003607e0 > > > > [1657077.998040] DR0: 0000000000000000 DR1: 0000000000000000 > > > > DR2: > > > > 0000000000000000 > > > > [1657077.998041] DR3: 0000000000000000 DR6: 00000000fffe0ff0 > > > > DR7: > > > > 0000000000000400 > > > > [1657077.998042] Call Trace: > > > > [1657077.998044] <IRQ> > > > > [1657077.998046] [<ffffffffc08a5527>] > > > > iscsi_itt_to_ctask+0x17/0x80 > > > > [libiscsi] > > > > [1657077.998050] [<ffffffffc05eefea>] iser_task_rsp+0xca/0x360 > > > > [ib_iser] > > > > [1657077.998061] [<ffffffffc0587fbb>] > > > > __ib_process_cq+0x6b/0xe0 > > > > [ib_core] > > > > [1657077.998066] [<ffffffffc0588122>] > > > > ib_poll_handler+0x22/0x80 > > > > [ib_core] > > > > [1657077.998070] [<ffffffff81358507>] > > > > irq_poll_softirq+0xc7/0x100 > > > > [1657077.998076] [<ffffffff81095195>] __do_softirq+0xf5/0x280 > > > > [1657077.998081] [<ffffffff816c4e8c>] call_softirq+0x1c/0x30 > > > > [1657077.998086] [<ffffffff8102d435>] do_softirq+0x65/0xa0 > > > > [1657077.998088] [<ffffffff81095515>] irq_exit+0x105/0x110 > > > > [1657077.998091] [<ffffffff816c61d6>] do_IRQ+0x56/0xf0 > > > > [1657077.998098] [<ffffffff816b837c>] > > > > common_interrupt+0x17c/0x17c > > > > [1657077.998099] <EOI> > > > > [1657077.998113] Code: ff ff ff eb a9 41 be 95 ff ff ff eb a1 > > > > 0f > > > > 1f > > > > 44 > > > > 00 00 55 48 89 e5 41 55 41 54 49 89 fc 53 89 f3 48 83 ec 10 c7 > > > > 45 > > > > d8 00 > > > > 00 00 00 <4c> 8b 6f 10 65 48 8b 04 25 28 00 00 00 48 89 45 e0 > > > > 31 > > > > c0 > > > > 83 > > > > fe > > > > [1657077.998116] RIP [<ffffffffc08a541e>] > > > > iscsi_verify_itt+0x1e/0x110 > > > > [libiscsi] > > > > [1657077.998116] RSP <ffff88beff403d78> > > > > [1657077.998117] CR2: 0000000000000010 > > > > crash> > > > > > > > > crash> bt > > > > PID: 41538 TASK: ffff88587ce38fd0 CPU: 20 COMMAND: "sh" > > > > #0 [ffff88beff403a18] machine_kexec at ffffffff8105ddeb > > > > #1 [ffff88beff403a78] __crash_kexec at ffffffff81109902 > > > > #2 [ffff88beff403b48] crash_kexec at ffffffff811099f0 > > > > #3 [ffff88beff403b60] oops_end at ffffffff816b97a8 > > > > #4 [ffff88beff403b88] no_context at ffffffff816a8c96 > > > > #5 [ffff88beff403bd8] __bad_area_nosemaphore at > > > > ffffffff816a8d2c > > > > #6 [ffff88beff403c20] bad_area_nosemaphore at > > > > ffffffff816a8e96 > > > > #7 [ffff88beff403c30] __do_page_fault at ffffffff816bc6be > > > > #8 [ffff88beff403c90] do_page_fault at ffffffff816bc865 > > > > #9 [ffff88beff403cc0] page_fault at ffffffff816b8788 > > > > [exception RIP: iscsi_verify_itt+30] > > > > RIP: ffffffffc08a541e RSP: ffff88beff403d78 RFLAGS: > > > > 00010286 > > > > RAX: 000000000000004c RBX: 00000000b0000036 RCX: > > > > 0000000000000002 > > > > RDX: 00000000000000cc RSI: 00000000b0000036 RDI: > > > > 0000000000000000 > > > > RBP: ffff88beff403da0 R8: 0000000040032a20 R9: > > > > ffff8896e4eaf91c > > > > R10: 0000000000000000 R11: 00007ffff7763ca0 R12: > > > > 0000000000000000 > > > > R13: ffff8896e4eaf9e4 R14: ffff8896e4eaf900 R15: > > > > 0000000000000000 > > > > ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0000 > > > > #10 [ffff88beff403da8] iscsi_itt_to_ctask at ffffffffc08a5527 > > > > [libiscsi] > > > > #11 [ffff88beff403dc8] iser_task_rsp at ffffffffc05eefea > > > > [ib_iser] > > > > #12 [ffff88beff403e10] __ib_process_cq at ffffffffc0587fbb > > > > [ib_core] > > > > #13 [ffff88beff403e50] ib_poll_handler at ffffffffc0588122 > > > > [ib_core] > > > > #14 [ffff88beff403e80] irq_poll_softirq at ffffffff81358507 > > > > #15 [ffff88beff403eb8] __do_softirq at ffffffff81095195 > > > > #16 [ffff88beff403f28] call_softirq at ffffffff816c4e8c > > > > #17 [ffff88beff403f40] do_softirq at ffffffff8102d435 > > > > #18 [ffff88beff403f60] irq_exit at ffffffff81095515 > > > > #19 [ffff88beff403f78] do_IRQ at ffffffff816c61d6 > > > > --- <IRQ stack> --- > > > > #20 [ffff884dd0af3f58] ret_from_intr at ffffffff816b837c > > > > RIP: 000000000041b866 RSP: 00007fffffffea28 RFLAGS: > > > > 00000206 > > > > RAX: 0000000000000000 RBX: 00007fffffffef53 RCX: > > > > 00000000006f1a70 > > > > RDX: 00000000006f1a70 RSI: 00000000006f1a90 RDI: > > > > 0000000000000000 > > > > RBP: 0000000000000002 R8: 0000000000000001 R9: > > > > 0000000000000020 > > > > R10: 0000000000000003 R11: 00007ffff7763ca0 R12: > > > > ffff88beff4061e8 > > > > R13: 00000000ffffffff R14: 0000000000000000 R15: > > > > 0000000000000063 > > > > ORIG_RAX: ffffffffffffffbb CS: 0033 SS: 002b > > > > > > > > crash> ps -p 41538 > > > > PID: 0 TASK: ffffffff81a0e480 CPU: 0 COMMAND: > > > > "swapper/0" > > > > PID: 1 TASK: ffff88012e4c8000 CPU: 7 COMMAND: > > > > "systemd" > > > > PID: 2345 TASK: ffff885ef5eb8fd0 CPU: 14 COMMAND: > > > > "zabbix_agentd" > > > > PID: 2349 TASK: ffff885efcbcaf70 CPU: 1 COMMAND: > > > > "zabbix_agentd" > > > > PID: 41538 TASK: ffff88587ce38fd0 CPU: 20 COMMAND: "sh" > > > > > > > > > > > > > > Don > > I misspoke about the kernel version, its 7.4 > > 3.10.0-693.34.1.el7_bz1582551.x86_64 > > Its the one we added the missing iscsi patches to but base is 7.4 > > So I will test with 7.5 > > > > Don, I had another look at this. > > Its not the SG_GAPS issue causing a memory registration error I > reported and we fixed in 7.5 from upstream. > > Which commit in 7.5 did we pull in for fix this from upstream. > > I think this is different and not yet fixed ?? > > [14556.614551] iser: iser_err_comp: memreg failure: memory management > operation error (6) vend_err 78 > [14556.666134] connection1:0: detected conn error (1011) > [14562.678414] mlx5_1:dump_cqe:262:(pid 0): dump error cqe > [14562.678529] mlx5_0:dump_cqe:262:(pid 0): dump error cqe > [14562.678530] 00000000 00000000 00000000 00000000 > [14562.678531] 00000000 00000000 00000000 00000000 > [14562.678531] 00000000 00000000 00000000 00000000 > [14562.678532] 00000000 08007806 25000344 34681cd2 > [14562.678535] iser: iser_err_comp: memreg failure: memory management > operation error (6) vend_err 78 > [14562.678544] connection1:0: detected conn error (1011) > [14562.679098] BUG: unable to handle kernel NULL pointer dereference > at > 0000000000000010 > [14562.679105] IP: [<ffffffffc088141e>] iscsi_verify_itt+0x1e/0x110 > [libiscsi] > [14562.679106] PGD 0 > [14562.679107] Oops: 0000 [#1] SMP > [14562.679134] Modules linked in: ip6table_filter ip6_tables > iptable_filter sctp_diag sctp tcp_diag udp_diag inet_diag unix_diag > af_packet_diag netlink_diag bnx2i cnic uio ip_vs nf_conntrack > oracleadvm(POE) oracleoks(POE) oracleasm(O) nfsv3 rpcsec_gss_krb5 > nfsv4 > dns_resolver nfs fscache dm_round_robin bonding rpcrdma ib_isert > iscsi_target_mod ib_iser libiscsi scsi_transport_iscsi ib_srpt > target_core_mod ib_srp scsi_transport_srp scsi_tgt ib_ipoib rdma_ucm > ib_ucm ib_uverbs ib_umad rdma_cm ib_cm iw_cm mlx5_ib ib_core xfs vfat > fat sb_edac edac_core intel_powerclamp coretemp intel_rapl iosf_mbi > kvm_intel kvm irqbypass crc32_pclmul ghash_clmulni_intel aesni_intel > lrw gf128mul glue_helper ablk_helper cryptd iTCO_wdt > iTCO_vendor_support ipmi_ssif pcspkr ipmi_si dm_multipath ioatdma > lpc_ich i2c_i801 sg hpilo > [14562.679152] hpwdt dca ipmi_devintf ipmi_msghandler pcc_cpufreq > shpchp wmi acpi_power_meter binfmt_misc nfsd auth_rpcgss nfs_acl > lockd > grace sunrpc ip_tables ext4 mbcache jbd2 sd_mod crc_t10dif > crct10dif_generic i2c_algo_bit drm_kms_helper syscopyarea sysfillrect > sysimgblt fb_sys_fops ttm bnx2x mlx5_core devlink mdio tg3(OE) > libcrc32c drm crct10dif_pclmul hpsa(OE) crct10dif_common ptp i2c_core > crc32c_intel scsi_transport_sas pps_core dm_mirror dm_region_hash > dm_log dm_mod > [14562.679154] CPU: 9 PID: 0 Comm: swapper/9 Tainted: > P OE - > ----------- 3.10.0-693.22.1.el7.x86_64 #1 > [14562.679155] Hardware name: HP ProLiant DL380 Gen9/ProLiant DL380 > Gen9, BIOS P89 05/21/2018 > [14562.679156] task: ffff8860aefaaf70 ti: ffff8860ae440000 task.ti: > ffff8860ae440000 > [14562.679158] RIP: 0010:[<ffffffffc088141e>] [<ffffffffc088141e>] > iscsi_verify_itt+0x1e/0x110 [libiscsi] > [14562.679159] RSP: 0018:ffff88beff2c3d78 EFLAGS: 00010286 > [14562.679160] RAX: 000000000000004c RBX: 00000000d0000041 RCX: > 0000000000000002 > [14562.679161] RDX: 00000000000000cc RSI: 00000000d0000041 RDI: > 0000000000000000 > [14562.679161] RBP: ffff88beff2c3da0 R08: 0000000040001038 R09: > ffff88ae496fe01c > [14562.679162] R10: 0000000000000000 R11: 7fffffffffffffff R12: > 0000000000000000 > [14562.679162] R13: ffff88ae496fe0e4 R14: ffff88ae496fe000 R15: > 0000000000000000 > [14562.679163] FS: 0000000000000000(0000) GS:ffff88beff2c0000(0000) > knlGS:0000000000000000 > [14562.679164] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 > [14562.679164] CR2: 0000000000000010 CR3: 000000beede48000 CR4: > 00000000003607e0 > [14562.679165] DR0: 0000000000000000 DR1: 0000000000000000 DR2: > 0000000000000000 > [14562.679166] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: > 0000000000000400 > [14562.679166] Call Trace: > [14562.679168] <IRQ> > [14562.679170] [<ffffffffc0881527>] iscsi_itt_to_ctask+0x17/0x80 > [libiscsi] > [14562.679173] [<ffffffffc069ffea>] iser_task_rsp+0xca/0x360 > [ib_iser] > [14562.679181] [<ffffffffc0924fbb>] __ib_process_cq+0x6b/0xe0 > [ib_core] Starts with the memreg failures crash> log | grep "iser: iser_err_comp: memreg failure" | wc -l 1237 Then the panic [14556.614551] iser: iser_err_comp: memreg failure: memory management operation error (6) vend_err 78 [14556.666134] connection1:0: detected conn error (1011) [14562.678414] mlx5_1:dump_cqe:262:(pid 0): dump error cqe [14562.678529] mlx5_0:dump_cqe:262:(pid 0): dump error cqe [14562.678530] 00000000 00000000 00000000 00000000 [14562.678531] 00000000 00000000 00000000 00000000 [14562.678531] 00000000 00000000 00000000 00000000 [14562.678532] 00000000 08007806 25000344 34681cd2 [14562.678535] iser: iser_err_comp: memreg failure: memory management operation error (6) vend_err 78 [14562.678544] connection1:0: detected conn error (1011) [14562.679098] BUG: unable to handle kernel NULL pointer dereference at 0000000000000010 [14562.679105] IP: [<ffffffffc088141e>] iscsi_verify_itt+0x1e/0x110 [libiscsi] crash> bt PID: 0 TASK: ffff8860aefaaf70 CPU: 9 COMMAND: "swapper/9" #0 [ffff88beff2c3a18] machine_kexec at ffffffff8105d77b #1 [ffff88beff2c3a78] __crash_kexec at ffffffff81108732 #2 [ffff88beff2c3b48] crash_kexec at ffffffff81108820 #3 [ffff88beff2c3b60] oops_end at ffffffff816b8778 #4 [ffff88beff2c3b88] no_context at ffffffff816a7c7a #5 [ffff88beff2c3bd8] __bad_area_nosemaphore at ffffffff816a7d10 #6 [ffff88beff2c3c20] bad_area_nosemaphore at ffffffff816a7e7a #7 [ffff88beff2c3c30] __do_page_fault at ffffffff816bb68e #8 [ffff88beff2c3c90] do_page_fault at ffffffff816bb835 #9 [ffff88beff2c3cc0] page_fault at ffffffff816b7768 [exception RIP: iscsi_verify_itt+30] RIP: ffffffffc088141e RSP: ffff88beff2c3d78 RFLAGS: 00010286 RAX: 000000000000004c RBX: 00000000d0000041 RCX: 0000000000000002 RDX: 00000000000000cc RSI: 00000000d0000041 RDI: 0000000000000000 RBP: ffff88beff2c3da0 R8: 0000000040001038 R9: ffff88ae496fe01c R10: 0000000000000000 R11: 7fffffffffffffff R12: 0000000000000000 R13: ffff88ae496fe0e4 R14: ffff88ae496fe000 R15: 0000000000000000 ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0018 #10 [ffff88beff2c3da8] iscsi_itt_to_ctask at ffffffffc0881527 [libiscsi] #11 [ffff88beff2c3dc8] iser_task_rsp at ffffffffc069ffea [ib_iser] #12 [ffff88beff2c3e10] __ib_process_cq at ffffffffc0924fbb [ib_core] #13 [ffff88beff2c3e50] ib_poll_handler at ffffffffc0925122 [ib_core] #14 [ffff88beff2c3e80] irq_poll_softirq at ffffffff813572b7 #15 [ffff88beff2c3eb8] __do_softirq at ffffffff81094035 #16 [ffff88beff2c3f28] call_softirq at ffffffff816c3afc #17 [ffff88beff2c3f40] do_softirq at ffffffff8102d435 #18 [ffff88beff2c3f60] irq_exit at ffffffff810943b5 #19 [ffff88beff2c3f78] do_IRQ at ffffffff816c4d96 --- <IRQ stack> --- #20 [ffff8860ae443db8] ret_from_intr at ffffffff816b7362 [exception RIP: cpuidle_enter_state+87] RIP: ffffffff81530b07 RSP: ffff8860ae443e60 RFLAGS: 00000202 RAX: 00000d3e7d729de6 RBX: ffff8860ae443e40 RCX: 0000000000000018 RDX: 0000000225c17d03 RSI: ffff8860ae443fd8 RDI: 00000d3e7d729de6 RBP: ffff8860ae443e88 R8: 000000000000016c R9: 000000000000001c R10: 0000000000000043 R11: 7fffffffffffffff R12: 0000000000000009 R13: ffff88beff2d39a0 R14: ffffffff810b77e5 R15: ffff8860ae443de0 ORIG_RAX: ffffffffffffff5d CS: 0010 SS: 0018 #21 [ffff8860ae443e90] cpuidle_idle_call at ffffffff81530c5e #22 [ffff8860ae443ed0] arch_cpu_idle at ffffffff81034f8e #23 [ffff8860ae443ee0] cpu_startup_entry at ffffffff810eb6da #24 [ffff8860ae443f28] start_secondary at ffffffff81052222 crash> dis -l iscsi_verify_itt+30 /usr/src/debug/kernel-3.10.0-693.22.1.el7/linux-3.10.0- 693.22.1.el7.x86_64/drivers/scsi/libiscsi.c: 1292 0xffffffffc088141e <iscsi_verify_itt+30>: mov 0x10(%rdi),%r13 crash> So fails here int iscsi_verify_itt(struct iscsi_conn *conn, itt_t itt) { struct iscsi_session *session = conn->session; **** conn- >session is invalid rdi had the struct iscsi_conn 0xffffffffc0881400 <iscsi_verify_itt>: nopl 0x0(%rax,%rax,1) [FTRACE NOP] 0xffffffffc0881405 <iscsi_verify_itt+5>: push %rbp 0xffffffffc0881406 <iscsi_verify_itt+6>: mov %rsp,%rbp 0xffffffffc0881409 <iscsi_verify_itt+9>: push %r13 0xffffffffc088140b <iscsi_verify_itt+11>: push %r12 0xffffffffc088140d <iscsi_verify_itt+13>: mov %rdi,%r12 0xffffffffc0881410 <iscsi_verify_itt+16>: push %rbx 0xffffffffc0881411 <iscsi_verify_itt+17>: mov %esi,%ebx 0xffffffffc0881413 <iscsi_verify_itt+19>: sub $0x10,%rsp 0xffffffffc0881417 <iscsi_verify_itt+23>: movl $0x0,-0x28(%rbp) 0xffffffffc088141e <iscsi_verify_itt+30>: mov 0x10(%rdi),%r13 RIP: ffffffffc088141e RSP: ffff88beff2c3d78 RFLAGS: 00010286 RAX: 000000000000004c RBX: 00000000d0000041 RCX: 0000000000000002 RDX: 00000000000000cc RSI: 00000000d0000041 RDI: 0000000000000000 RBP: ffff88beff2c3da0 R8: 0000000040001038 R9: ffff88ae496fe01c R10: 0000000000000000 R11: 7fffffffffffffff R12: 0000000000000000 R13: ffff88ae496fe0e4 R14: ffff88ae496fe000 R15: 0000000000000000 Both RDI and R12 are null, offset by 10 get the bad address So we have a race somehow that trashes the conn pointer under load. The load clearly is seeing resource issues and repeatedly failing the memory registration. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
