Hide Forgot
Description of problem: running a guest ubuntu14.04 img in a rhel 7.1 host causes kernel crash Version-Release number of selected component (if applicable): How reproducible: it can be reproduced in a specific physical host while restart the guest vm。 once restart the vm ,the pyhsical machine get crashed。 call trace: [ 43.547542] BUG: unable to handle kernel paging request at 00000000a2a09579 [ 43.547609] IP: [<00000000a2a09579>] 0xa2a09578 [ 43.547655] PGD 0 [ 43.547677] Oops: 0010 [#1] SMP [ 43.547709] Modules linked in: nbd(OF-) nf_conntrack_ipv6 nf_defrag_ipv6 ip6table_raw xt_mac xt_conntrack xt_physdev xt_CT iptable_mangle xt_comment iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack act_police cls_basic sch_ingress vhost_net macvtap macvlan tun iptable_raw veth bridge stp llc ebtable_filter ebtables ip6table_filter ip6_tables iptable_filter openvswitch gre vxlan ip_tunnel dm_mirror dm_region_hash dm_log dm_mod intel_powerclamp coretemp intel_rapl kvm_intel kvm crc32_pclmul ghash_clmulni_intel aesni_intel iTCO_wdt lrw gf128mul glue_helper ablk_helper cryptd iTCO_vendor_support dcdbas ipmi_devintf sg mei_me sb_edac pcspkr mei edac_core shpchp lpc_ich mfd_core ipmi_si wmi ipmi_msghandler acpi_power_meter nfsd auth_rpcgss nfs_acl lockd sunrpc ip_tables xfs [ 43.548405] libcrc32c sd_mod crc_t10dif mgag200 syscopyarea sysfillrect sysimgblt drm_kms_helper ttm crct10dif_pclmul crct10dif_common ixgbe crc32c_intel drm igb ahci libahci mdio ptp i2c_algo_bit libata pps_core i2c_core megaraid_sas dca [last unloaded: nbd] [ 43.548627] CPU: 34 PID: 4581 Comm: qemu-kvm Tainted: GF O-------------- 3.10.0-229.20.1.el7.x86_64 #1 [ 43.548696] Hardware name: Dell Inc. PowerEdge R630/0CNCJW, BIOS 1.3.6 06/03/2015 [ 43.548748] task: ffff881fcb932d80 ti: ffff881fb20a0000 task.ti: ffff881fb20a0000 [ 43.548799] RIP: 0010:[<00000000a2a09579>] [<00000000a2a09579>] 0xa2a09578 [ 43.548855] RSP: 0018:ffff881fb20a3ca8 EFLAGS: 00010002 [ 43.548892] RAX: 000000007b605ec6 RBX: 00000001316b4d60 RCX: 00000000000006e0 [ 43.548941] RDX: 000000000000000b RSI: 000000007b605ec6 RDI: 00000000000006e0 [ 43.548989] RBP: ffff88003e403e90 R08: 0000000000000004 R09: 0000000000000000 [ 43.549037] R10: 0000000000000004 R11: 0000000000000005 R12: 0000000084caf3cb [ 43.549086] R13: 0000000000000000 R14: 00000004aab72b00 R15: 0000000425ec18e2 [ 43.549135] FS: 00007fd2a652f700(0000) GS:ffff881fff620000(0000) knlGS:0000000000000000 [ 43.549190] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 43.549230] CR2: 00000000a2a09579 CR3: 0000003fa42b0000 CR4: 00000000001427e0 [ 43.549278] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 43.549327] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 [ 43.549374] Stack: [ 43.549391] ffff883fa0e30000 ffff883fa0e30000 ffff881fb20a3d28 0000000000006c14 [ 43.549457] 0000000000000000 ffff883fa0e30000 00000000b20a3cf0 ffffffffa06a65d7 [ 43.549524] ffff883fa0e30000 ffff881fb20a3d28 00000000b654bfd4 ffff883fa0e30000 [ 43.549591] Call Trace: [ 43.549634] [<ffffffffa06a65d7>] ? kvm_set_shared_msr+0x57/0x80 [kvm] [ 43.549696] [<ffffffffa06ad8b3>] ? vcpu_enter_guest+0x5e3/0xd70 [kvm] [ 43.549748] [<ffffffff810980f6>] ? finish_wait+0x56/0x70 [ 43.549796] [<ffffffffa06999db>] ? kvm_vcpu_block+0xab/0xd0 [kvm] [ 43.549842] [<ffffffff81098240>] ? wake_up_bit+0x30/0x30 [ 43.549896] [<ffffffffa06b2218>] ? kvm_arch_vcpu_ioctl_run+0x208/0x480 [kvm] [ 43.551615] [<ffffffffa069c32a>] ? kvm_vcpu_ioctl+0x2aa/0x580 [kvm] [ 43.553325] [<ffffffff811da035>] ? do_vfs_ioctl+0x2e5/0x4c0 [ 43.555046] [<ffffffffa06a6674>] ? kvm_on_user_return+0x74/0x80 [kvm] [ 43.556754] [<ffffffff811da2b1>] ? SyS_ioctl+0xa1/0xc0 [ 43.558439] [<ffffffff81614409>] ? system_call_fastpath+0x16/0x1b [ 43.560103] Code: Bad RIP value. [ 43.561781] RIP [<00000000a2a09579>] 0xa2a09578 [ 43.563489] RSP <ffff881fb20a3ca8> [ 43.565199] CR2: 00000000a2a09579
I missed this one, sorry. I didn't manage to reproduce ... does it still happen? Thanks.
We had two fixes for the shared msr paths that can be seen in stack trace, but the original report is not clear enough on the reproducer to determine whether this was a very specific manifestation of those bugs. Please reopen if you still happen to hit this.