System panic in update_group_capacity() due to a divide error

This document (7023575) is provided subject to the disclaimer at the end of this document.

Environment

SUSE Linux Enterprise Server 12 Service Pack 3 (SLES 12 SP3)

Situation

Systems with SLES12 SP3 kernel 4.4.143-94.47.1-default or older may experience kernel panics when running benchmarks. The stack trace of the failing task looks like:

PID: 0      TASK: ffff880cc4488340  CPU: 32  COMMAND: "swapper/32"
 #0 [ffff8817fcd03980] machine_kexec at ffffffff81058cfc
 #1 [ffff8817fcd039d0] __crash_kexec at ffffffff8111738a
 #2 [ffff8817fcd03a90] crash_kexec at ffffffff8111745c
 #3 [ffff8817fcd03aa0] oops_end at ffffffff8101a5d4
 #4 [ffff8817fcd03ac0] do_error_trap at ffffffff810181c0
 #5 [ffff8817fcd03b70] divide_error at ffffffff81616dae
    [exception RIP: update_group_capacity+0x169]
    RIP: ffffffff810bca59  RSP: ffff8817fcd03c20  RFLAGS: 00010247
    RAX: 00000000007468aa  RBX: ffff880cc34e9000  RCX: 0000000000000000
    RDX: 0000000000000000  RSI: 00000000e2329b00  RDI: 000000000000024d
    RBP: 0000000000000000   R8: ffff8817fcd00000   R9: 0000000000016440
    R10: ffff880cc34e7640  R11: ffff8817fcd16d00  R12: ffff880cc34e7640
    R13: 0000000000000020  R14: ffff8817fcd03c88  R15: 0000000000000000
    ORIG_RAX: ffffffffffffffff  CS: 0010  SS: 0018
 #6 [ffff8817fcd03c40] update_sd_lb_stats at ffffffff810bcba8
 #7 [ffff8817fcd03d08] find_busiest_group at ffffffff810bd04d
 #8 [ffff8817fcd03df8] load_balance at ffffffff810bd640
 #9 [ffff8817fcd03ee8] rebalance_domains at ffffffff810be3ff
#10 [ffff8817fcd03f40] __do_softirq at ffffffff810852bc
#11 [ffff8817fcd03fa0] irq_exit at ffffffff8108578a
#12 [ffff8817fcd03fb0] reschedule_interrupt at ffffffff816167d9
--- <IRQ stack> ---
#13 [ffff880cc448fde0] reschedule_interrupt at ffffffff816167d9
    RIP: 0000000000000018  RSP: 0000000000000000  RFLAGS: ffff880cc4490000
    RAX: ffffe8fffdb0f710  RBX: 000097d88dd1a446  RCX: 0000000000000008
    RDX: 0000000000000eca  RSI: 00000000ffffffff  RDI: 0000000000000349
    RBP: ffffffff810f6ad8   R8: 0000000000000003   R9: 0000000000000020
    R10: 000097d88d95fff8  R11: 0000000000000020  R12: ffff8817fcd11ce0
    R13: ffff8817fcd118c0  R14: ffffffff8132fd74  R15: ffffffff810fcac7
    ORIG_RAX: ffff8817fcd16440  CS: 6aebe2  SS: ffffffffffffff02

Resolution

A possible fix for this panic went into kernel 4.4.155-94.50.1-default.

Cause

The panic is caused by an overflow of a kernel variable.

Additional Information

Disclaimer

This Support Knowledgebase provides a valuable tool for SUSE customers and parties interested in our products and solutions to acquire information, ideas and learn from one another. Materials are provided for informational, personal or non-commercial use within your organization and are presented "AS IS" WITHOUT WARRANTY OF ANY KIND.