[lkml]   [2012]   [Apr]   [5]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
    SubjectRe: CPU Hotplug rework

    Here is my attempt at a summary of the discussion.

    Srivatsa, I left out the preempt_disable() pieces, but would be happy
    to add them in when you let me know what you are thinking to do for
    de-stop_machine()ing CPU hotplug.

    Thanx, Paul


    CPU-hotplug work breakout:

    1. Read and understand the current generic code.
    Srivatsa Bhat has done this, as have Paul E. McKenney and
    Peter Zijlstra to a lesser extent.

    2. Read and understand the architecture-specific code, looking
    for opportunities to consolidate additional function into
    core code.

    a. Carry out any indicated consolidation.

    b. Convert all architectures to make use of the
    consolidated implementation.

    Not started. Low priority from a big.LITTLE perspective.

    3. Address the current kthread creation/teardown/migration
    performance issues. (More details below.)

    Highest priority from a big.LITTLE perspective.

    4. Wean CPU-hotplug offlining from stop_machine().
    (More details below.)

    Moderate priority from a big.LITTLE perspective.


    1. Evaluate approaches. Approaches currently under
    consideration include:

    a. Park the kthreads rather than tearing them down or
    migrating them. RCU currently takes this sort of
    approach. Note that RCU currently relies on both
    preempt_disable() and local_bh_disable() blocking the
    current CPU from going offline.

    b. Allow in-kernel kthreads to avoid the delay
    required to work around a bug in old versions of
    bash. (This bug is a failure to expect receiving
    a SIGCHILD signal corresponding to a child
    created by a fork() system call that has not yet

    This might be implemented using an additional
    CLONE_ flag. This should allow kthreads to
    be created and torn down much more quickly.

    c. Have some other TBD way to "freeze" a kthread.
    (As in "your clever idea here".)

    2. Implement the chosen approach or approaches. (Different
    kernel subsystems might have different constraints, possibly
    requiring different kthread handling.)


    1. CPU_DYING notifier fixes needed as of 3.2:

    o vfp_hotplug(): I believe that this works as-is.
    o s390_nohz_notify(): I believe that this works as-is.
    o x86_pmu_notifier(): I believe that this works as-is.
    o perf_ibs_cpu_notifier(): I don't know enough about
    APIC to say.
    o tboot_cpu_callback(): I believe that this works as-is,
    but this one returns NOTIFY_BAD to a CPU_DYING notifier,
    which is badness. But it looks like that case is a
    "cannot happen" case. Still needs to be fixed.
    o clockevents_notify(): This one acquires a global lock,
    so it should be safe as-is.
    o console_cpu_notify(): This one takes the same action
    CPU_UP_CANCELLED that it does for CPU_DYING, so it
    should be OK.
    * rcu_cpu_notify(): This one needs adjustment as noted
    above, but nothing major. Patch has been posted,
    probably needs a bit of debugging.
    o migration_call(): I defer to Peter on this one.
    It looks to me like it is written to handle other
    CPUs, but...
    * workqueue_cpu_callback(): Might need help, does a
    non-atomic OR.
    o kvm_cpu_hotplug(): Uses a global spinlock, so should
    be OK as-is.

    2. Evaluate designs for stop_machine()-free CPU hotplug.
    Implement the chosen design. An outline for a particular
    design is shown below, but the actual design might be
    quite different.

    3. Fix issues with CPU Hotplug callback registration. Currently
    there is no totally-race-free way to register callbacks and do
    setup for already online cpus.

    Srivatsa had posted an incomplete patchset some time ago
    regarding this, which gives an idea of the direction he had
    in mind.

    4. There is a mismatch between the code and the documentation around
    the difference between [un/register]_hotcpu_notifier and
    [un/register]_cpu_notifier. And I remember seeing several places
    in the code that uses them inconsistently. Not terribly important,
    but good to fix it up while we are at it.

    5. There was another thread where stuff related to CPU hotplug had
    been discussed. It had exposed some new challenges to CPU hotplug,
    if we were to support asynchronous smp booting.

    6. If preempt_disable() no longer blocks CPU offlining, then
    uses of preempt_disable() in the kernel need to be inspected
    to see which are relying on blocking offlining, and any
    identified will need adjustment.


    1. preempt_disable() or something similarly lightweight and
    unconditional must block removal of any CPU that was
    in cpu_online_map at the start of the "critical section".
    (I will identify these as hotplug read-side critical sections.)

    I don't believe that there is any prohibition against a CPU
    appearing suddenly, but some auditing would be required to
    confirm this. But see below.

    2. A subsystem not involved in the CPU-hotplug process must be able
    to test if a CPU is online and be guaranteed that this test
    remains valid (the CPU remains fully functional) for the duration
    of the hotplug read-side critical section.

    3. If a subsystem needs to operate on all currently online CPUs,
    then it must participate in the CPU-hotplug process. My
    belief is that if some code needs to test whether a CPU is
    present, and needs an "offline" indication to persist, then
    that code's subsystem must participate in CPU-hotplug operations.

    4. There must be a way to register/unregister for CPU-hotplug events.
    This is currently cpu_notifier(), register_cpu_notifier(),
    and unregister_cpu_notifier().

    n-1. CPU-hotplug operations should be reasonably fast. A few
    milliseconds is OK, multiple seconds not so much.

    n. (Your additional constraints here.)


    a. Maintain the cpu_online_map, as currently, but the meaning
    of a set bit is that the CPU is fully functional. If there
    is any service that the CPU no longer offers, its bit is

    b. Continue to use preempt_enable()/preempt_disable() to mark
    hotplug read-side critical sections.

    c. Instead of using __stop_machine(), use a per-CPU variable that
    is checked in the idle loop. Possibly another TIF_ bit.

    d. The CPU notifiers are like today, except that CPU_DYING() is
    invoked by the CPU after it sees that its per-CPU variable
    telling it to go offline. As today, the CPU_DYING notifiers
    are invoked with interrupts disabled, but other CPUs are still
    running. Of course, the CPU_DYING notifiers need to be audited
    and repaired. There are fewer than 20 of them, so not so bad.
    RCU's is an easy fix: Just re-introduce locking and the global
    RCU callback orphanage. My guesses for the others at the end.

    e. Getting rid of __stop_machine() means that the final step of the
    CPU going offline will no longer be seen as atomic by other CPUs.
    This will require more careful tracking of dependencies among
    different subsystems. The required tracking can be reduced
    by invoking notifiers in registration order for CPU-online
    operations and invoking them in the reverse of registration
    order for CPU-offline operations.

    For example, the scheduler uses RCU. If notifiers are invoked in
    the same order for all CPU-hotplug operations, then on CPU-offline
    operations, during the time between when RCU's notifier is invoked
    and when the scheduler's notifier is invoked, the scheduler must
    deal with a CPU on which RCU isn't working. (RCU currently
    works around this by allowing a one-jiffy time period after
    notification when it still pays attention to the CPU.)

    In contrast, if notifiers are invoked in reverse-registration
    order for CPU-offline operations, then any time the scheduler
    sees a CPU as online, RCU also is treating it as online.

    f. There will be some circular dependencies. For example, the
    scheduler uses RCU, but in some configurations, RCU also uses
    kthreads. These dependencies must be handled on a case-by-case
    basis. For example, the scheduler could invoke an RCU API
    to tell RCU when to shut down its per-CPU kthreads and when
    to start them up. Or RCU could deal with its kthreads in the
    CPU_DOWN_PREPARE and CPU_ONLINE notifiers. Either way, RCU
    needs to correctly handle the interval when it cannot use
    kthreads on a given CPU that it is still handling, for example,
    by switching to running the RCU core code in softirq context.

    g. Most subsystems participating in CPU-hotplug operations will need
    to keep their own copy of CPU online/offline state. For example,
    RCU uses the ->qsmaskinit fields in the rcu_node structure for
    this purpose.

    h. So CPU-offline handling looks something like the following:

    i. Acquire the hotplug mutex.

    ii. Invoke the CPU_DOWN_PREPARE notifiers. If there
    are objections, invoke the CPU_DOWN_FAILED notifiers
    and return an error.

    iii. Clear the CPU's bit in cpu_online_map.

    iv. Invoke synchronize_sched() to ensure that all future hotplug
    read-side critical sections ignore the outgoing CPU.

    v. Set a per-CPU variable telling the CPU to take itself
    offline. There would need to be something here to
    help the CPU get to idle quickly, possibly requiring
    another round of notifiers. CPU_DOWN?

    vi. When the dying CPU gets to the idle loop, it invokes the
    CPU_DYING notifiers and updates its per-CPU variable to
    indicate that it is ready to die. It then spins in a
    tight loop (or does some other architecture-specified
    operation to wait to be turned off).

    Note that there is no need for RCU to guess how long the
    CPU might be executing RCU read-side critical sections.

    vii. When the task doing the offline operation sees the
    updated per-CPU variable, it calls __cpu_die().

    viii. The CPU_DEAD notifiers are invoked.

    ix. Theeck_for_tasks() function is invoked.

    x. Release the hotplug mutex.

    xi. Invoke the CPU_POST_DEAD notifiers.

    i. I do not believe that the CPU-offline handling needs to change

     \ /
      Last update: 2012-04-06 01:09    [W:0.045 / U:32.696 seconds]
    ©2003-2017 Jasper Spaans. hosted at Digital OceanAdvertise on this site