lkml.org 
[lkml]   [2015]   [May]   [8]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    Patch in this message
    /
    From
    Subject[PATCH 6/6] nohz: add dataplane_debug boot flag
    Date
    This flag simplifies debugging of NO_HZ_FULL kernels when processes
    are running in PR_DATAPLANE_QUIESCE mode. Such processes should
    get no interrupts from the kernel, and if they do, when this boot
    flag is specified a kernel stack dump on the console is generated.

    It's possible to use ftrace to simply detect whether a dataplane core
    has unexpectedly entered the kernel. But what this boot flag does
    is allow the kernel to provide better diagnostics, e.g. by reporting
    in the IPI-generating code what remote core and context is preparing
    to deliver an interrupt to a dataplane core.

    It may be worth considering other ways to generate useful debugging
    output rather than console spew, but for now that is simple and direct.

    Signed-off-by: Chris Metcalf <cmetcalf@ezchip.com>
    ---
    Documentation/kernel-parameters.txt | 6 ++++++
    arch/tile/mm/homecache.c | 5 ++++-
    include/linux/tick.h | 2 ++
    kernel/irq_work.c | 4 +++-
    kernel/sched/core.c | 18 ++++++++++++++++++
    kernel/signal.c | 5 +++++
    kernel/smp.c | 4 ++++
    kernel/softirq.c | 1 +
    8 files changed, 43 insertions(+), 2 deletions(-)

    diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
    index f6befa9855c1..5c5af5258e17 100644
    --- a/Documentation/kernel-parameters.txt
    +++ b/Documentation/kernel-parameters.txt
    @@ -794,6 +794,12 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
    dasd= [HW,NET]
    See header of drivers/s390/block/dasd_devmap.c.

    + dataplane_debug [KNL]
    + In kernels built with CONFIG_NO_HZ_FULL and booted
    + in nohz_full= mode, this setting will generate console
    + backtraces when the kernel is about to interrupt a
    + task that has requested PR_DATAPLANE_QUIESCE.
    +
    db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
    (one device per port)
    Format: <port#>,<type>
    diff --git a/arch/tile/mm/homecache.c b/arch/tile/mm/homecache.c
    index 40ca30a9fee3..dd5ec7eca9a8 100644
    --- a/arch/tile/mm/homecache.c
    +++ b/arch/tile/mm/homecache.c
    @@ -31,6 +31,7 @@
    #include <linux/smp.h>
    #include <linux/module.h>
    #include <linux/hugetlb.h>
    +#include <linux/tick.h>

    #include <asm/page.h>
    #include <asm/sections.h>
    @@ -83,8 +84,10 @@ static void hv_flush_update(const struct cpumask *cache_cpumask,
    * Don't bother to update atomically; losing a count
    * here is not that critical.
    */
    - for_each_cpu(cpu, &mask)
    + for_each_cpu(cpu, &mask) {
    ++per_cpu(irq_stat, cpu).irq_hv_flush_count;
    + tick_nohz_dataplane_debug(cpu);
    + }
    }

    /*
    diff --git a/include/linux/tick.h b/include/linux/tick.h
    index d191cda9b71a..4610cdf0f972 100644
    --- a/include/linux/tick.h
    +++ b/include/linux/tick.h
    @@ -147,6 +147,7 @@ extern void tick_nohz_full_kick_cpu(int cpu);
    extern void tick_nohz_full_kick_all(void);
    extern void __tick_nohz_task_switch(struct task_struct *tsk);
    extern void tick_nohz_dataplane_enter(void);
    +extern void tick_nohz_dataplane_debug(int cpu);
    #else
    static inline bool tick_nohz_full_enabled(void) { return false; }
    static inline bool tick_nohz_full_cpu(int cpu) { return false; }
    @@ -157,6 +158,7 @@ static inline void tick_nohz_full_kick_all(void) { }
    static inline void __tick_nohz_task_switch(struct task_struct *tsk) { }
    static inline bool tick_nohz_is_dataplane(void) { return false; }
    static inline void tick_nohz_dataplane_enter(void) { }
    +static inline void tick_nohz_dataplane_debug(int cpu) { }
    #endif

    static inline bool is_housekeeping_cpu(int cpu)
    diff --git a/kernel/irq_work.c b/kernel/irq_work.c
    index cbf9fb899d92..0adc53c4e899 100644
    --- a/kernel/irq_work.c
    +++ b/kernel/irq_work.c
    @@ -75,8 +75,10 @@ bool irq_work_queue_on(struct irq_work *work, int cpu)
    if (!irq_work_claim(work))
    return false;

    - if (llist_add(&work->llnode, &per_cpu(raised_list, cpu)))
    + if (llist_add(&work->llnode, &per_cpu(raised_list, cpu))) {
    + tick_nohz_dataplane_debug(cpu);
    arch_send_call_function_single_ipi(cpu);
    + }

    return true;
    }
    diff --git a/kernel/sched/core.c b/kernel/sched/core.c
    index f9123a82cbb6..202fab0c41cb 100644
    --- a/kernel/sched/core.c
    +++ b/kernel/sched/core.c
    @@ -719,6 +719,24 @@ bool sched_can_stop_tick(void)

    return true;
    }
    +
    +/* Enable debugging of any interrupts of dataplane cores. */
    +static int dataplane_debug;
    +static int __init dataplane_debug_func(char *str)
    +{
    + dataplane_debug = true;
    + return 1;
    +}
    +__setup("dataplane_debug", dataplane_debug_func);
    +
    +void tick_nohz_dataplane_debug(int cpu)
    +{
    + if (dataplane_debug && tick_nohz_full_cpu(cpu) &&
    + (cpu_curr(cpu)->dataplane_flags & PR_DATAPLANE_QUIESCE)) {
    + pr_err("Interrupt detected for dataplane cpu %d\n", cpu);
    + dump_stack();
    + }
    +}
    #endif /* CONFIG_NO_HZ_FULL */

    void sched_avg_update(struct rq *rq)
    diff --git a/kernel/signal.c b/kernel/signal.c
    index d51c5ddd855c..ebc552cafff5 100644
    --- a/kernel/signal.c
    +++ b/kernel/signal.c
    @@ -689,6 +689,11 @@ int dequeue_signal(struct task_struct *tsk, sigset_t *mask, siginfo_t *info)
    */
    void signal_wake_up_state(struct task_struct *t, unsigned int state)
    {
    +#ifdef CONFIG_NO_HZ_FULL
    + /* If the task is being killed, don't complain about dataplane. */
    + if (state & TASK_WAKEKILL)
    + t->dataplane_flags = 0;
    +#endif
    set_tsk_thread_flag(t, TIF_SIGPENDING);
    /*
    * TASK_WAKEKILL also means wake it up in the stopped/traced/killable
    diff --git a/kernel/smp.c b/kernel/smp.c
    index 07854477c164..9518fc80321b 100644
    --- a/kernel/smp.c
    +++ b/kernel/smp.c
    @@ -14,6 +14,7 @@
    #include <linux/smp.h>
    #include <linux/cpu.h>
    #include <linux/sched.h>
    +#include <linux/tick.h>

    #include "smpboot.h"

    @@ -178,6 +179,7 @@ static int generic_exec_single(int cpu, struct call_single_data *csd,
    * locking and barrier primitives. Generic code isn't really
    * equipped to do the right thing...
    */
    + tick_nohz_dataplane_debug(cpu);
    if (llist_add(&csd->llist, &per_cpu(call_single_queue, cpu)))
    arch_send_call_function_single_ipi(cpu);

    @@ -457,6 +459,8 @@ void smp_call_function_many(const struct cpumask *mask,
    }

    /* Send a message to all CPUs in the map */
    + for_each_cpu(cpu, cfd->cpumask)
    + tick_nohz_dataplane_debug(cpu);
    arch_send_call_function_ipi_mask(cfd->cpumask);

    if (wait) {
    diff --git a/kernel/softirq.c b/kernel/softirq.c
    index bc9406337f82..eeacabf08ca6 100644
    --- a/kernel/softirq.c
    +++ b/kernel/softirq.c
    @@ -394,6 +394,7 @@ void irq_exit(void)
    WARN_ON_ONCE(!irqs_disabled());
    #endif

    + tick_nohz_dataplane_debug(smp_processor_id());
    account_irq_exit_time(current);
    preempt_count_sub(HARDIRQ_OFFSET);
    if (!in_interrupt() && local_softirq_pending())
    --
    2.1.2


    \
     
     \ /
      Last update: 2015-05-08 20:21    [W:4.044 / U:0.684 seconds]
    ©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site