On Tue, Aug 14, 2018 at 10:06:18AM -0700, Paul E. McKenney wrote:
> On Tue, Aug 14, 2018 at 12:49:45PM -0400, Steven Rostedt wrote:
> > On Tue, 14 Aug 2018 09:24:48 -0700
> > "Paul E. McKenney" <paul...@linux.vnet.ibm.com> wrote:
> >
> > > Event tracing is moving to SRCU in order to take advantage of the fact
> > > that SRCU may be safely used from idle and even offline CPUs. However,
> > > event tracing can invoke call_srcu() very early in the boot process,
> > > even before workqueue_init_early() is invoked (let alone rcu_init()).
> > > Therefore, call_srcu()'s attempts to queue work fail miserably.
> > >
> > > This commit therefore detects this situation, and refrains from attempting
> > > to queue work before rcu_init() time, but does everything else that it
> > > would have done, and in addition, adds the srcu_struct to a global list.
> > > The rcu_init() function now invokes a new srcu_init() function, which
> > > is empty if CONFIG_SRCU=n. Otherwise, srcu_init() queues work for
> > > each srcu_struct on the list. This all happens early enough in boot
> > > that there is but a single CPU with interrupts disabled, which allows
> > > synchronization to be dispensed with.
> > >
> > > Of course, the queued work won't actually be invoked until after
> > > workqueue_init() is invoked, which happens shortly after the scheduler
> > > is up and running. This means that although call_srcu() may be invoked
> > > any time after per-CPU variables have been set up, there is still a very
> > > narrow window when synchronize_srcu() won't work, and this window
> > > extends from the time that the scheduler starts until the time that
> > > workqueue_init() returns. This can be fixed in a manner similar to
> > > the fix for synchronize_rcu_expedited() and friends, but until someone
> > > actually needs to use synchronize_srcu() during this window, this fix
> > > is added churn for no benefit.
> > >
> > > Finally, note that Tree SRCU's new srcu_init() function invokes
> > > queue_work() rather than the queue_delayed_work() function that is invoked
> > > post-boot. The reason is that queue_delayed_work() will (as you would
> > > expect) post a timer, and timers have not yet been initialized. So use
> > > of queue_delayed_work() avoids the complaints about use of uninitialized
> >
> > You mean "So use of queue_work() avoids .." ?
>
> Indeed I do! Fixed.
>
> > > spinlocks that would otherwise result. Besides, delay is in any case
> > > provide by the aforementioned fact that the queued work won't actually
> > > be invoked until after the scheduler is up and running.
> > >
> > > Requested-by: Steven Rostedt <rost...@goodmis.org>
> > > Signed-off-by: Paul E. McKenney <paul...@linux.vnet.ibm.com>
> > >
> > > diff --git a/include/linux/srcutiny.h b/include/linux/srcutiny.h
> > > index f41d2fb09f87..2b5c0822e683 100644
> > > --- a/include/linux/srcutiny.h
> > > +++ b/include/linux/srcutiny.h
> > > @@ -36,6 +36,7 @@ struct srcu_struct {
> > > struct rcu_head *srcu_cb_head; /* Pending callbacks: Head. */
> > > struct rcu_head **srcu_cb_tail; /* Pending callbacks: Tail. */
> > > struct work_struct srcu_work; /* For driving grace periods. */
> > > + struct list_head srcu_boot_entry; /* Early-boot callbacks. */
> >
> > I really don't like increasing the size of a structure for a field that
> > is hardly ever used.
> >
> > Is there a way we could make a union, or reuse one of the other fields,
> > as we know that synchronize_srcu() can't be used yet (and if it is,
> > either warn, or just make it a nop). And when we call srcu_init() and
> > remove the srcu_struct from the list, we can then initialize whatever
> > we used as the temporary boot up list field.
>
> I will take a look. If nothing else, I could union it with the
> struct work_struct, since it cannot be used that early anyway. ;-)
Not so much!!! The problem is that the srcu_struct needs to be
initialized differently depending on whether it is used before or after
start_kernel()'s call to rcu_init(). Before, it needs to be initialized
as a list_head, after as a work_struct. But the type of initialization
is determined not by the time of initialization but rather by the time
of first use. So it looks like reusing work_struct's list_head makes
more sense.
> Or I could just use the work_struct that is already inside the struct
> work_struct. Tejun, would you be OK with that?
I am creating a separate patch that eliminates the boot-time-only
->srcu_boot_entry field to allow the decisions to be made separately.
Thanx, Paul
> For whatever it is worth, synchronize_srcu() is perfectly legal way
> early in boot, at least as early as call_srcu(). The reason is that
> until the scheduler starts, synchronize_srcu() is a no-op.
>
> > srcu_init is called when we are still running only one CPU correct?
>
> Yes, single CPU interrupts disabled.
>
> > > #ifdef CONFIG_DEBUG_LOCK_ALLOC
> > > struct lockdep_map dep_map;
> > > #endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
> > > @@ -48,6 +49,7 @@ void srcu_drive_gp(struct work_struct *wp);
> > > .srcu_wq = __SWAIT_QUEUE_HEAD_INITIALIZER(name.srcu_wq), \
> > > .srcu_cb_tail = &name.srcu_cb_head, \
> > > .srcu_work = __WORK_INITIALIZER(name.srcu_work, srcu_drive_gp), \
> > > + .srcu_boot_entry = LIST_HEAD_INIT(name.srcu_boot_entry), \
> > > __SRCU_DEP_MAP_INIT(name) \
> > > }
> > >
> > > diff --git a/include/linux/srcutree.h b/include/linux/srcutree.h
> > > index 745d4ca4dd50..86ad97111315 100644
> > > --- a/include/linux/srcutree.h
> > > +++ b/include/linux/srcutree.h
> > > @@ -94,6 +94,7 @@ struct srcu_struct {
> > > /* callback for the barrier */
> > > /* operation. */
> > > struct delayed_work work;
> > > + struct list_head srcu_boot_entry; /* Early-boot callbacks. */
> > > #ifdef CONFIG_DEBUG_LOCK_ALLOC
> > > struct lockdep_map dep_map;
> > > #endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
> > > @@ -105,12 +106,13 @@ struct srcu_struct {
> > > #define SRCU_STATE_SCAN2 2
> > >
> > > #define __SRCU_STRUCT_INIT(name, pcpu_name)
> > > \
> > > - { \
> > > - .sda = &pcpu_name, \
> > > - .lock = __SPIN_LOCK_UNLOCKED(name.lock), \
> > > - .srcu_gp_seq_needed = 0 - 1, \
> > > - __SRCU_DEP_MAP_INIT(name) \
> > > - }
> > > +{
> > > \
> > > + .sda = &pcpu_name, \
> > > + .lock = __SPIN_LOCK_UNLOCKED(name.lock), \
> > > + .srcu_gp_seq_needed = 0 - 1, \
> >
> > Interesting initialization of -1. This was there before, but still
> > interesting none the less.
>
> If I recall correctly, this subterfuge suppresses compiler complaints
> about initializing an unsigned long with a negative number. :-/
>
> Thanx, Paul
>
> > > + .srcu_boot_entry = LIST_HEAD_INIT(name.srcu_boot_entry), \
> > > + __SRCU_DEP_MAP_INIT(name) \
> > > +}
> > >
> > >
> >
> > -- Steve
> >
