kernel 4.4.2: kvm_irq_delivery_to_api / rwsem_down_read_failed

DISCUSSION Stefan Priebe 2 participants 4 messages Feb 20, 2016 – Feb 22, 2016 KVM

Share & tools

Stefan Priebe

author

2016-02-20 10:43 UTC #0 lore raw

Hi,

while testing Kernel 4.4.2 and starting 20 Qemu 2.4.1 virtual machines. I got those traces and a load of 500 on those system. I was only abler to recover by sysrq-trigger.

All traces:

INFO: task pvedaemon worke:7470 blocked for more than 120 seconds.

Not tainted 4.4.2+1-ph #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. pvedaemon worke D ffff88239c367ca0 0 7470 7468 0x00080000 ffff88239c367ca0 ffff8840a6232500 ffff8823ed83a500 ffff88239c367c90 ffff88239c368000 ffff8845f5f070e8 ffff8845f5f07100 0000000000000000 00007ffc73b48e58 ffff88239c367cc0 ffffffffb66a4d89 ffff88239c367cf0

Call Trace:
[<ffffffffb66a4d89>] schedule+0x39/0x80
[<ffffffffb66a7447>] rwsem_down_read_failed+0xc7/0x120
[<ffffffffb63cb594>] call_rwsem_down_read_failed+0x14/0x30
[<ffffffffb66a6af7>] ? down_read+0x17/0x20
[<ffffffffb617b99e>] __access_remote_vm+0x3e/0x1c0
[<ffffffffb63cb594>] ? call_rwsem_down_read_failed+0x14/0x30
[<ffffffffb6181d5f>] access_remote_vm+0x1f/0x30
[<ffffffffb623212e>] proc_pid_cmdline_read+0x16e/0x4f0
[<ffffffffb611a4cc>] ? acct_account_cputime+0x1c/0x20
[<ffffffffb61c8348>] __vfs_read+0x18/0x40
[<ffffffffb61c94fe>] vfs_read+0x8e/0x140
[<ffffffffb61c95ff>] SyS_read+0x4f/0xa0
[<ffffffffb66a892e>] entry_SYSCALL_64_fastpath+0x12/0x71
INFO: task pvestatd:7633 blocked for more than 120 seconds.

Not tainted 4.4.2+1-ph #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. pvestatd D ffff88239f16fd40 0 7633 1 0x00080000 ffff88239f16fd40 ffff8824e76a8000 ffff8823e5fc2500 ffff8823e5fc2500 ffff88239f170000 ffff8845f5f070e8 ffff8845f5f07100 ffff8845f5f07080 000000000341bf10 ffff88239f16fd60 ffffffffb66a4d89 024000d000000058

Call Trace:
[<ffffffffb66a4d89>] schedule+0x39/0x80
[<ffffffffb66a7447>] rwsem_down_read_failed+0xc7/0x120
[<ffffffffb63cb594>] call_rwsem_down_read_failed+0x14/0x30
[<ffffffffb66a6af7>] ? down_read+0x17/0x20
[<ffffffffb623206c>] proc_pid_cmdline_read+0xac/0x4f0
[<ffffffffb611a4cc>] ? acct_account_cputime+0x1c/0x20
[<ffffffffb60b1f23>] ? account_user_time+0x73/0x80
[<ffffffffb60b249e>] ? vtime_account_user+0x4e/0x70
[<ffffffffb61c8348>] __vfs_read+0x18/0x40
[<ffffffffb61c94fe>] vfs_read+0x8e/0x140
[<ffffffffb61c95ff>] SyS_read+0x4f/0xa0
[<ffffffffb66a892e>] entry_SYSCALL_64_fastpath+0x12/0x71
INFO: task kvm:11766 blocked for more than 120 seconds.

Not tainted 4.4.2+1-ph #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. kvm D ffff88452a2d3dd0 0 11766 1 0x00080000
ffff88452a2d3dd0 ffff880166c74a00 ffff8845b7354a00 ffffffffb617fc8e ffff88452a2d4000 ffff8845f5f070e8 ffff8845f5f07100 ffff88452a2d3f58 ffff8845b7354a00 ffff88452a2d3df0 ffffffffb66a4d89 00007fa807abbf80

Call Trace:
[<ffffffffb617fc8e>] ? __handle_mm_fault+0xd1e/0x1260
[<ffffffffb66a4d89>] schedule+0x39/0x80
[<ffffffffb66a7447>] rwsem_down_read_failed+0xc7/0x120
[<ffffffffb63cb594>] call_rwsem_down_read_failed+0x14/0x30
[<ffffffffb66a6af7>] ? down_read+0x17/0x20
[<ffffffffb604f477>] __do_page_fault+0x2b7/0x380
[<ffffffffb60b1f23>] ? account_user_time+0x73/0x80
[<ffffffffb60b249e>] ? vtime_account_user+0x4e/0x70
[<ffffffffb604f5a7>] do_page_fault+0x37/0x90
[<ffffffffb66aa7b8>] page_fault+0x28/0x30
INFO: task kvm:11824 blocked for more than 120 seconds.

Not tainted 4.4.2+1-ph #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. kvm D ffff8840a867faa0 0 11824 1 0x00080000
ffff8840a867faa0 ffff8845866a4a00 ffff8840a6232500 0000000000000001 ffff8840a8680000 ffff8845f5f070e8 ffff8845f5f07100 ffff8840a867fc0e 0000000000000000 ffff8840a867fac0 ffffffffb66a4d89 ffffffffc0606a06

Call Trace:
[<ffffffffb66a4d89>] schedule+0x39/0x80
[<ffffffffc0606a06>] ? kvm_irq_delivery_to_apic+0x56/0x220 [kvm]
[<ffffffffb66a7447>] rwsem_down_read_failed+0xc7/0x120
[<ffffffffb63cb594>] call_rwsem_down_read_failed+0x14/0x30
[<ffffffffb66a6af7>] ? down_read+0x17/0x20
[<ffffffffc05d1480>] kvm_host_page_size+0x60/0xa0 [kvm]
[<ffffffffc05ea9bc>] mapping_level+0x5c/0x130 [kvm]
[<ffffffffc05f1b1b>] tdp_page_fault+0x9b/0x260 [kvm]
[<ffffffffc05d70ed>] ? kernel_pio+0x2d/0x40 [kvm]
[<ffffffffc05eba21>] kvm_mmu_page_fault+0x31/0x120 [kvm]
[<ffffffffc0678db4>] handle_ept_violation+0xa4/0x170 [kvm_intel]
[<ffffffffc067fd07>] vmx_handle_exit+0x257/0x490 [kvm_intel]
[<ffffffffb60b2081>] ? __vtime_account_system+0x31/0x40
[<ffffffffc05e662f>] vcpu_enter_guest+0x6af/0xff0 [kvm]
[<ffffffffc06034ad>] ? kvm_apic_local_deliver+0x5d/0x60 [kvm]
[<ffffffffc05e8564>] kvm_arch_vcpu_ioctl_run+0xc4/0x3c0 [kvm]
[<ffffffffc05cf844>] kvm_vcpu_ioctl+0x324/0x5d0 [kvm]
[<ffffffffb611a4cc>] ? acct_account_cputime+0x1c/0x20
[<ffffffffb60b1f23>] ? account_user_time+0x73/0x80
[<ffffffffb61da203>] do_vfs_ioctl+0x83/0x4e0
[<ffffffffb600261f>] ? enter_from_user_mode+0x1f/0x50
[<ffffffffb6002711>] ? syscall_trace_enter_phase1+0xc1/0x110
[<ffffffffb61da6ac>] SyS_ioctl+0x4c/0x80
[<ffffffffb66a892e>] entry_SYSCALL_64_fastpath+0x12/0x71
INFO: task kvm:11825 blocked for more than 120 seconds.

Not tainted 4.4.2+1-ph #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. kvm D ffff88458d6a3aa0 0 11825 1 0x00080002
ffff88458d6a3aa0 ffff880167302500 ffff8840a6234a00 0000000000000001 ffff88458d6a4000 ffff8845f5f070e8 ffff8845f5f07100 ffff88458d6a3c0e 0000000000000000 ffff88458d6a3ac0 ffffffffb66a4d89 ffffffffc0606a06

Call Trace:
[<ffffffffb66a4d89>] schedule+0x39/0x80
[<ffffffffc0606a06>] ? kvm_irq_delivery_to_apic+0x56/0x220 [kvm]
[<ffffffffb66a7447>] rwsem_down_read_failed+0xc7/0x120
[<ffffffffb63cb594>] call_rwsem_down_read_failed+0x14/0x30
[<ffffffffb66a6af7>] ? down_read+0x17/0x20
[<ffffffffc05d1480>] kvm_host_page_size+0x60/0xa0 [kvm]
[<ffffffffc05ea9bc>] mapping_level+0x5c/0x130 [kvm]
[<ffffffffc05f1b1b>] tdp_page_fault+0x9b/0x260 [kvm]
[<ffffffffc05eba21>] kvm_mmu_page_fault+0x31/0x120 [kvm]
[<ffffffffc0678db4>] handle_ept_violation+0xa4/0x170 [kvm_intel]
[<ffffffffc067fd07>] vmx_handle_exit+0x257/0x490 [kvm_intel]
[<ffffffffb60b2081>] ? __vtime_account_system+0x31/0x40
[<ffffffffc05e662f>] vcpu_enter_guest+0x6af/0xff0 [kvm]
[<ffffffffc06034ad>] ? kvm_apic_local_deliver+0x5d/0x60 [kvm]
[<ffffffffc05e8564>] kvm_arch_vcpu_ioctl_run+0xc4/0x3c0 [kvm]
[<ffffffffc05cf844>] kvm_vcpu_ioctl+0x324/0x5d0 [kvm]
[<ffffffffb611a4cc>] ? acct_account_cputime+0x1c/0x20
[<ffffffffb60b1f23>] ? account_user_time+0x73/0x80
[<ffffffffb61da203>] do_vfs_ioctl+0x83/0x4e0
[<ffffffffb600261f>] ? enter_from_user_mode+0x1f/0x50
[<ffffffffb6002711>] ? syscall_trace_enter_phase1+0xc1/0x110
[<ffffffffb61da6ac>] SyS_ioctl+0x4c/0x80
[<ffffffffb66a892e>] entry_SYSCALL_64_fastpath+0x12/0x71
INFO: task kvm:14910 blocked for more than 120 seconds.

Not tainted 4.4.2+1-ph #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. kvm D ffff8838aee4fdd0 0 14910 1 0x00080000
ffff8838aee4fdd0 ffff882279b04a00 ffff883a20a5ca00 ffffffffb617facf ffff8838aee50000 ffff8845f5f070e8 ffff8845f5f07100 ffff8838aee4ff58 ffff883a20a5ca00 ffff8838aee4fdf0 ffffffffb66a4d89 ffff880000000040

Call Trace:
[<ffffffffb617facf>] ? __handle_mm_fault+0xb5f/0x1260
[<ffffffffb66a4d89>] schedule+0x39/0x80
[<ffffffffb66a7447>] rwsem_down_read_failed+0xc7/0x120
[<ffffffffb63cb594>] call_rwsem_down_read_failed+0x14/0x30
[<ffffffffb66a6af7>] ? down_read+0x17/0x20
[<ffffffffb604f477>] __do_page_fault+0x2b7/0x380
[<ffffffffb60b1f23>] ? account_user_time+0x73/0x80
[<ffffffffb60b249e>] ? vtime_account_user+0x4e/0x70
[<ffffffffb604f5a7>] do_page_fault+0x37/0x90
[<ffffffffb66aa7b8>] page_fault+0x28/0x30
INFO: task kvm:14912 blocked for more than 120 seconds.

Not tainted 4.4.2+1-ph #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. kvm D ffff8845683efdf8 0 14912 1 0x00080000
ffff8845683efdf8 ffffffffb6c0f4c0 ffff883998edca00 ffff883998edca00 ffff8845683f0000 ffff8845f5f07100 ffff8845f5f070e8 ffffffff00000000 ffffffff0000000a ffff8845683efe18 ffffffffb66a4d89 0000000000000000

Call Trace:
[<ffffffffb66a4d89>] schedule+0x39/0x80
[<ffffffffb66a7237>] rwsem_down_write_failed+0x1b7/0x300
[<ffffffffb63cb5c3>] call_rwsem_down_write_failed+0x13/0x20
[<ffffffffb66a6b24>] ? down_write+0x24/0x40
[<ffffffffb6188901>] SyS_mprotect+0xc1/0x210
[<ffffffffb66a892e>] entry_SYSCALL_64_fastpath+0x12/0x71
INFO: task kvm:15177 blocked for more than 120 seconds.

Not tainted 4.4.2+1-ph #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. kvm D ffff8845eb73fdd0 0 15177 1 0x00080000
ffff8845eb73fdd0 ffff8824e76aca00 ffff883994e20000 ffffffffb617facf ffff8845eb740000 ffff8845f5f070e8 ffff8845f5f07100 ffff8845eb73ff58 ffff883994e20000 ffff8845eb73fdf0 ffffffffb66a4d89 ffff880000000038

Call Trace:
[<ffffffffb617facf>] ? __handle_mm_fault+0xb5f/0x1260
[<ffffffffb66a4d89>] schedule+0x39/0x80
[<ffffffffb66a7447>] rwsem_down_read_failed+0xc7/0x120
[<ffffffffb60cbc11>] ? rwsem_wake+0x71/0xb0
[<ffffffffb63cb594>] call_rwsem_down_read_failed+0x14/0x30
[<ffffffffb66a6af7>] ? down_read+0x17/0x20
[<ffffffffb604f477>] __do_page_fault+0x2b7/0x380
[<ffffffffb60b1f23>] ? account_user_time+0x73/0x80
[<ffffffffb60b249e>] ? vtime_account_user+0x4e/0x70
[<ffffffffb604f5a7>] do_page_fault+0x37/0x90
[<ffffffffb66aa7b8>] page_fault+0x28/0x30
INFO: task iotop:14292 blocked for more than 120 seconds.

Not tainted 4.4.2+1-ph #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. iotop D ffff88239130bca0 0 14292 14290 0x00080000 ffff88239130bca0 ffff8824e7670000 ffff88228bf70000 ffff88239130bc90 ffff88239130c000 ffff8845f5f070e8 ffff8845f5f07100 0000000000000000 00007ffc73b48e58 ffff88239130bcc0 ffffffffb66a4d89 ffff88239130bcf0

Call Trace:
[<ffffffffb66a4d89>] schedule+0x39/0x80
[<ffffffffb66a7447>] rwsem_down_read_failed+0xc7/0x120
[<ffffffffb63cb594>] call_rwsem_down_read_failed+0x14/0x30
[<ffffffffb66a6af7>] ? down_read+0x17/0x20
[<ffffffffb617b99e>] __access_remote_vm+0x3e/0x1c0
[<ffffffffb63cb594>] ? call_rwsem_down_read_failed+0x14/0x30
[<ffffffffb6181d5f>] access_remote_vm+0x1f/0x30
[<ffffffffb623212e>] proc_pid_cmdline_read+0x16e/0x4f0
[<ffffffffb611a4cc>] ? acct_account_cputime+0x1c/0x20
[<ffffffffb61c8348>] __vfs_read+0x18/0x40
[<ffffffffb61c94fe>] vfs_read+0x8e/0x140
[<ffffffffb61c95ff>] SyS_read+0x4f/0xa0
[<ffffffffb66a892e>] entry_SYSCALL_64_fastpath+0x12/0x71
INFO: task top:14293 blocked for more than 120 seconds.

Not tainted 4.4.2+1-ph #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. top D ffff8845ed873ca0 0 14293 14291 0x00080000 ffff8845ed873ca0 ffff8823ed83a500 ffff8836ab6dca00 ffff8845ed873c90 ffff8845ed874000 ffff8845f5f070e8 ffff8845f5f07100 0000000000000000 00007ffc73b48e58 ffff8845ed873cc0 ffffffffb66a4d89 ffff8845ed873cf0

Call Trace:
[<ffffffffb66a4d89>] schedule+0x39/0x80
[<ffffffffb66a7447>] rwsem_down_read_failed+0xc7/0x120
[<ffffffffb63cb594>] call_rwsem_down_read_failed+0x14/0x30
[<ffffffffb66a6af7>] ? down_read+0x17/0x20
[<ffffffffb617b99e>] __access_remote_vm+0x3e/0x1c0
[<ffffffffb63cb594>] ? call_rwsem_down_read_failed+0x14/0x30
[<ffffffffb6181d5f>] access_remote_vm+0x1f/0x30
[<ffffffffb623212e>] proc_pid_cmdline_read+0x16e/0x4f0
[<ffffffffb611a4cc>] ? acct_account_cputime+0x1c/0x20
[<ffffffffb61c8348>] __vfs_read+0x18/0x40
[<ffffffffb61c94fe>] vfs_read+0x8e/0x140
[<ffffffffb61c95ff>] SyS_read+0x4f/0xa0
[<ffffffffb66a892e>] entry_SYSCALL_64_fastpath+0x12/0x71

vmbr0: port 32(tap111i0) entered disabled state
vmbr0: port 32(tap111i0) entered disabled state

Greets,
Stefan

Paolo Bonzini

maintainer

2016-02-22 17:36 UTC #1 lore raw

Quoted 5 lines · Stefan Priebe

Hi,
while testing Kernel 4.4.2 and starting 20 Qemu 2.4.1 virtual machines. I got those traces and a load of 500 on those system. I was only abler to recover by sysrq-trigger.

It seems like something happening at the VM level. A task took the mm semaphore and hung everyone else. Difficult to debug without a core (and without knowing who held the semaphore). Sorry.

Paolo

Quoted 43 lines

All traces:
INFO: task pvedaemon worke:7470 blocked for more than 120 seconds.
Not tainted 4.4.2+1-ph #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. pvedaemon worke D ffff88239c367ca0 0 7470 7468 0x00080000 ffff88239c367ca0 ffff8840a6232500 ffff8823ed83a500 ffff88239c367c90 ffff88239c368000 ffff8845f5f070e8 ffff8845f5f07100 0000000000000000 00007ffc73b48e58 ffff88239c367cc0 ffffffffb66a4d89 ffff88239c367cf0
Call Trace:
[<ffffffffb66a4d89>] schedule+0x39/0x80
[<ffffffffb66a7447>] rwsem_down_read_failed+0xc7/0x120
[<ffffffffb63cb594>] call_rwsem_down_read_failed+0x14/0x30
[<ffffffffb66a6af7>] ? down_read+0x17/0x20
[<ffffffffb617b99e>] __access_remote_vm+0x3e/0x1c0
[<ffffffffb63cb594>] ? call_rwsem_down_read_failed+0x14/0x30
[<ffffffffb6181d5f>] access_remote_vm+0x1f/0x30
[<ffffffffb623212e>] proc_pid_cmdline_read+0x16e/0x4f0
[<ffffffffb611a4cc>] ? acct_account_cputime+0x1c/0x20
[<ffffffffb61c8348>] __vfs_read+0x18/0x40
[<ffffffffb61c94fe>] vfs_read+0x8e/0x140
[<ffffffffb61c95ff>] SyS_read+0x4f/0xa0
[<ffffffffb66a892e>] entry_SYSCALL_64_fastpath+0x12/0x71
INFO: task pvestatd:7633 blocked for more than 120 seconds.
Not tainted 4.4.2+1-ph #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. pvestatd D ffff88239f16fd40 0 7633 1 0x00080000 ffff88239f16fd40 ffff8824e76a8000 ffff8823e5fc2500 ffff8823e5fc2500 ffff88239f170000 ffff8845f5f070e8 ffff8845f5f07100 ffff8845f5f07080 000000000341bf10 ffff88239f16fd60 ffffffffb66a4d89 024000d000000058
Call Trace:
[<ffffffffb66a4d89>] schedule+0x39/0x80
[<ffffffffb66a7447>] rwsem_down_read_failed+0xc7/0x120
[<ffffffffb63cb594>] call_rwsem_down_read_failed+0x14/0x30
[<ffffffffb66a6af7>] ? down_read+0x17/0x20
[<ffffffffb623206c>] proc_pid_cmdline_read+0xac/0x4f0
[<ffffffffb611a4cc>] ? acct_account_cputime+0x1c/0x20
[<ffffffffb60b1f23>] ? account_user_time+0x73/0x80
[<ffffffffb60b249e>] ? vtime_account_user+0x4e/0x70
[<ffffffffb61c8348>] __vfs_read+0x18/0x40
[<ffffffffb61c94fe>] vfs_read+0x8e/0x140
[<ffffffffb61c95ff>] SyS_read+0x4f/0xa0
[<ffffffffb66a892e>] entry_SYSCALL_64_fastpath+0x12/0x71

Stefan Priebe

author

2016-02-22 19:34 UTC #2 lore raw

Am 22.02.2016 um 18:36 schrieb Paolo Bonzini:

Quoted 12 lines · Stefan Priebe

Quoted 5 lines
Hi,
while testing Kernel 4.4.2 and starting 20 Qemu 2.4.1 virtual machines. I got those traces and a load of 500 on those system. I was only abler to recover by sysrq-trigger.
It seems like something happening at the VM level. A task took the mm semaphore and hung everyone else. Difficult to debug without a core (and without knowing who held the semaphore). Sorry.

OK thank you anyway. Is there anything i can do if this happens again?

Stefan

Quoted 46 lines

Paolo
Quoted 43 lines
All traces:
INFO: task pvedaemon worke:7470 blocked for more than 120 seconds.
Not tainted 4.4.2+1-ph #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. pvedaemon worke D ffff88239c367ca0 0 7470 7468 0x00080000 ffff88239c367ca0 ffff8840a6232500 ffff8823ed83a500 ffff88239c367c90 ffff88239c368000 ffff8845f5f070e8 ffff8845f5f07100 0000000000000000 00007ffc73b48e58 ffff88239c367cc0 ffffffffb66a4d89 ffff88239c367cf0
Call Trace:
[<ffffffffb66a4d89>] schedule+0x39/0x80
[<ffffffffb66a7447>] rwsem_down_read_failed+0xc7/0x120
[<ffffffffb63cb594>] call_rwsem_down_read_failed+0x14/0x30
[<ffffffffb66a6af7>] ? down_read+0x17/0x20
[<ffffffffb617b99e>] __access_remote_vm+0x3e/0x1c0
[<ffffffffb63cb594>] ? call_rwsem_down_read_failed+0x14/0x30
[<ffffffffb6181d5f>] access_remote_vm+0x1f/0x30
[<ffffffffb623212e>] proc_pid_cmdline_read+0x16e/0x4f0
[<ffffffffb611a4cc>] ? acct_account_cputime+0x1c/0x20
[<ffffffffb61c8348>] __vfs_read+0x18/0x40
[<ffffffffb61c94fe>] vfs_read+0x8e/0x140
[<ffffffffb61c95ff>] SyS_read+0x4f/0xa0
[<ffffffffb66a892e>] entry_SYSCALL_64_fastpath+0x12/0x71
INFO: task pvestatd:7633 blocked for more than 120 seconds.
Not tainted 4.4.2+1-ph #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. pvestatd D ffff88239f16fd40 0 7633 1 0x00080000 ffff88239f16fd40 ffff8824e76a8000 ffff8823e5fc2500 ffff8823e5fc2500 ffff88239f170000 ffff8845f5f070e8 ffff8845f5f07100 ffff8845f5f07080 000000000341bf10 ffff88239f16fd60 ffffffffb66a4d89 024000d000000058
Call Trace:
[<ffffffffb66a4d89>] schedule+0x39/0x80
[<ffffffffb66a7447>] rwsem_down_read_failed+0xc7/0x120
[<ffffffffb63cb594>] call_rwsem_down_read_failed+0x14/0x30
[<ffffffffb66a6af7>] ? down_read+0x17/0x20
[<ffffffffb623206c>] proc_pid_cmdline_read+0xac/0x4f0
[<ffffffffb611a4cc>] ? acct_account_cputime+0x1c/0x20
[<ffffffffb60b1f23>] ? account_user_time+0x73/0x80
[<ffffffffb60b249e>] ? vtime_account_user+0x4e/0x70
[<ffffffffb61c8348>] __vfs_read+0x18/0x40
[<ffffffffb61c94fe>] vfs_read+0x8e/0x140
[<ffffffffb61c95ff>] SyS_read+0x4f/0xa0
[<ffffffffb66a892e>] entry_SYSCALL_64_fastpath+0x12/0x71

Paolo Bonzini

maintainer

2016-02-22 19:44 UTC #3 lore raw

----- Original Message -----

Quoted 16 lines

From: "Stefan Priebe" <[email protected]>
To: "Paolo Bonzini" <[email protected]>, "qemu-devel" <[email protected]>, [email protected]
Sent: Monday, February 22, 2016 8:35:41 PM
Subject: Re: kernel 4.4.2: kvm_irq_delivery_to_api / rwsem_down_read_failed
Am 22.02.2016 um 18:36 schrieb Paolo Bonzini:
Quoted 9 lines · Stefan Priebe
Quoted 5 lines
Hi,
while testing Kernel 4.4.2 and starting 20 Qemu 2.4.1 virtual machines. I got those traces and a load of 500 on those system. I was only abler to recover by sysrq-trigger.

Quoted 5 lines

Quoted 3 lines
It seems like something happening at the VM level. A task took the mm semaphore and hung everyone else. Difficult to debug without a core (and without knowing who held the semaphore). Sorry.
OK thank you anyway. Is there anything i can do if this happens again?

Try grabbing a vmcore with sysrq-c, if you have kdump configured.

Paolo