Re: PRREMPT RCU stalls with 9edfbfed3f544a78 ("sched/core: Rework rq->clock update skips")

[Mike Galbraith]

On Mon, 2015-06-29 at 18:37 +0100, Mark Rutland wrote:
> Hi,
> 
> Recently (since ~v4.0) I've been seeing new RCU stall warnings when
> running hackbench (and cpu-affine cyclictest instances) on
> CONFIG_PREEMPT=y kernels, both arm64 and x86_64:

FWIW, I'm unable to reproduce on my Haswell box w. preempt config.

homer: # grep RCU config.preempt
# RCU Subsystem
CONFIG_PREEMPT_RCU=y
CONFIG_RCU_EXPERT=y
CONFIG_SRCU=y
# CONFIG_TASKS_RCU is not set
CONFIG_RCU_STALL_COMMON=y
CONFIG_RCU_USER_QS=y
CONFIG_RCU_FANOUT=64
CONFIG_RCU_FANOUT_LEAF=16
# CONFIG_RCU_FAST_NO_HZ is not set
# CONFIG_TREE_RCU_TRACE is not set
# CONFIG_RCU_BOOST is not set
CONFIG_RCU_KTHREAD_PRIO=0
CONFIG_RCU_NOCB_CPU=y
CONFIG_RCU_NOCB_CPU_NONE=y
# CONFIG_RCU_NOCB_CPU_ZERO is not set
# CONFIG_RCU_NOCB_CPU_ALL is not set
# CONFIG_RCU_EXPEDITE_BOOT is not set
# RCU Debugging
# CONFIG_PROVE_RCU is not set
# CONFIG_SPARSE_RCU_POINTER is not set
# CONFIG_RCU_TORTURE_TEST is not set
CONFIG_RCU_CPU_STALL_TIMEOUT=60
# CONFIG_RCU_CPU_STALL_INFO is not set
# CONFIG_RCU_TRACE is not set
# CONFIG_RCU_EQS_DEBUG is not set

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/