Hi Thomas,
On Wed, Mar 18, 2020 at 09:43:13PM +0100, Thomas Gleixner wrote:
> From: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
>
> completion uses a wait_queue_head_t to enqueue waiters.
>
> wait_queue_head_t contains a spinlock_t to protect the list of waiters
> which excludes it from being used in truly atomic context on a PREEMPT_RT
> enabled kernel.
>
> The spinlock in the wait queue head cannot be replaced by a raw_spinlock
> because:
>
> - wait queues can have custom wakeup callbacks, which acquire other
> spinlock_t locks and have potentially long execution times
Cool, makes sense.
> - wake_up() walks an unbounded number of list entries during the wake up
> and may wake an unbounded number of waiters.
Just to clarify here, wake_up() will really wake up just 1 waiter if all the
waiters on the queue are exclusive right? So in such scenario at least, the
"unbounded number of waiters" would not be an issue if everything waiting was
exclusive and waitqueue with wake_up() was used. Please correct me if I'm
wrong about that though.
So the main reasons to avoid waitqueue in favor of swait (as you mentioned)
would be the sleep-while-atomic issue in truly atomic context on RT, and the
fact that callbacks can take a long time.
>
> For simplicity and performance reasons complete() should be usable on
> PREEMPT_RT enabled kernels.
>
> completions do not use custom wakeup callbacks and are usually single
> waiter, except for a few corner cases.
>
> Replace the wait queue in the completion with a simple wait queue (swait),
> which uses a raw_spinlock_t for protecting the waiter list and therefore is
> safe to use inside truly atomic regions on PREEMPT_RT.
>
> There is no semantical or functional change:
>
> - completions use the exclusive wait mode which is what swait provides
>
> - complete() wakes one exclusive waiter
>
> - complete_all() wakes all waiters while holding the lock which protects
> the wait queue against newly incoming waiters. The conversion to swait
> preserves this behaviour.
>
> complete_all() might cause unbound latencies with a large number of waiters
> being woken at once, but most complete_all() usage sites are either in
> testing or initialization code or have only a really small number of
> concurrent waiters which for now does not cause a latency problem. Keep it
> simple for now.
>
> The fixup of the warning check in the USB gadget driver is just a straight
> forward conversion of the lockless waiter check from one waitqueue type to
> the other.
>
> Signed-off-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> Cc: Arnd Bergmann <arnd@xxxxxxxx>
Reviewed-by: Joel Fernandes (Google) <joel@xxxxxxxxxxxxxxxxx>
thanks,
- Joel
> ---
> V2: Split out the orinoco and usb gadget parts and amended change log
> ---
> drivers/usb/gadget/function/f_fs.c | 2 +-
> include/linux/completion.h | 8 ++++----
> kernel/sched/completion.c | 36 +++++++++++++++++++-----------------
> 3 files changed, 24 insertions(+), 22 deletions(-)
>
> --- a/drivers/usb/gadget/function/f_fs.c
> +++ b/drivers/usb/gadget/function/f_fs.c
> @@ -1703,7 +1703,7 @@ static void ffs_data_put(struct ffs_data
> pr_info("%s(): freeing\n", __func__);
> ffs_data_clear(ffs);
> BUG_ON(waitqueue_active(&ffs->ev.waitq) ||
> - waitqueue_active(&ffs->ep0req_completion.wait) ||
> + swait_active(&ffs->ep0req_completion.wait) ||
> waitqueue_active(&ffs->wait));
> destroy_workqueue(ffs->io_completion_wq);
> kfree(ffs->dev_name);
> --- a/include/linux/completion.h
> +++ b/include/linux/completion.h
> @@ -9,7 +9,7 @@
> * See kernel/sched/completion.c for details.
> */
>
> -#include <linux/wait.h>
> +#include <linux/swait.h>
>
> /*
> * struct completion - structure used to maintain state for a "completion"
> @@ -25,7 +25,7 @@
> */
> struct completion {
> unsigned int done;
> - wait_queue_head_t wait;
> + struct swait_queue_head wait;
> };
>
> #define init_completion_map(x, m) __init_completion(x)
> @@ -34,7 +34,7 @@ static inline void complete_acquire(stru
> static inline void complete_release(struct completion *x) {}
>
> #define COMPLETION_INITIALIZER(work) \
> - { 0, __WAIT_QUEUE_HEAD_INITIALIZER((work).wait) }
> + { 0, __SWAIT_QUEUE_HEAD_INITIALIZER((work).wait) }
>
> #define COMPLETION_INITIALIZER_ONSTACK_MAP(work, map) \
> (*({ init_completion_map(&(work), &(map)); &(work); }))
> @@ -85,7 +85,7 @@ static inline void complete_release(stru
> static inline void __init_completion(struct completion *x)
> {
> x->done = 0;
> - init_waitqueue_head(&x->wait);
> + init_swait_queue_head(&x->wait);
> }
>
> /**
> --- a/kernel/sched/completion.c
> +++ b/kernel/sched/completion.c
> @@ -29,12 +29,12 @@ void complete(struct completion *x)
> {
> unsigned long flags;
>
> - spin_lock_irqsave(&x->wait.lock, flags);
> + raw_spin_lock_irqsave(&x->wait.lock, flags);
>
> if (x->done != UINT_MAX)
> x->done++;
> - __wake_up_locked(&x->wait, TASK_NORMAL, 1);
> - spin_unlock_irqrestore(&x->wait.lock, flags);
> + swake_up_locked(&x->wait);
> + raw_spin_unlock_irqrestore(&x->wait.lock, flags);
> }
> EXPORT_SYMBOL(complete);
>
> @@ -58,10 +58,12 @@ void complete_all(struct completion *x)
> {
> unsigned long flags;
>
> - spin_lock_irqsave(&x->wait.lock, flags);
> + WARN_ON(irqs_disabled());
> +
> + raw_spin_lock_irqsave(&x->wait.lock, flags);
> x->done = UINT_MAX;
> - __wake_up_locked(&x->wait, TASK_NORMAL, 0);
> - spin_unlock_irqrestore(&x->wait.lock, flags);
> + swake_up_all_locked(&x->wait);
> + raw_spin_unlock_irqrestore(&x->wait.lock, flags);
> }
> EXPORT_SYMBOL(complete_all);
>
> @@ -70,20 +72,20 @@ do_wait_for_common(struct completion *x,
> long (*action)(long), long timeout, int state)
> {
> if (!x->done) {
> - DECLARE_WAITQUEUE(wait, current);
> + DECLARE_SWAITQUEUE(wait);
>
> - __add_wait_queue_entry_tail_exclusive(&x->wait, &wait);
> do {
> if (signal_pending_state(state, current)) {
> timeout = -ERESTARTSYS;
> break;
> }
> + __prepare_to_swait(&x->wait, &wait);
> __set_current_state(state);
> - spin_unlock_irq(&x->wait.lock);
> + raw_spin_unlock_irq(&x->wait.lock);
> timeout = action(timeout);
> - spin_lock_irq(&x->wait.lock);
> + raw_spin_lock_irq(&x->wait.lock);
> } while (!x->done && timeout);
> - __remove_wait_queue(&x->wait, &wait);
> + __finish_swait(&x->wait, &wait);
> if (!x->done)
> return timeout;
> }
> @@ -100,9 +102,9 @@ static inline long __sched
>
> complete_acquire(x);
>
> - spin_lock_irq(&x->wait.lock);
> + raw_spin_lock_irq(&x->wait.lock);
> timeout = do_wait_for_common(x, action, timeout, state);
> - spin_unlock_irq(&x->wait.lock);
> + raw_spin_unlock_irq(&x->wait.lock);
>
> complete_release(x);
>
> @@ -291,12 +293,12 @@ bool try_wait_for_completion(struct comp
> if (!READ_ONCE(x->done))
> return false;
>
> - spin_lock_irqsave(&x->wait.lock, flags);
> + raw_spin_lock_irqsave(&x->wait.lock, flags);
> if (!x->done)
> ret = false;
> else if (x->done != UINT_MAX)
> x->done--;
> - spin_unlock_irqrestore(&x->wait.lock, flags);
> + raw_spin_unlock_irqrestore(&x->wait.lock, flags);
> return ret;
> }
> EXPORT_SYMBOL(try_wait_for_completion);
> @@ -322,8 +324,8 @@ bool completion_done(struct completion *
> * otherwise we can end up freeing the completion before complete()
> * is done referencing it.
> */
> - spin_lock_irqsave(&x->wait.lock, flags);
> - spin_unlock_irqrestore(&x->wait.lock, flags);
> + raw_spin_lock_irqsave(&x->wait.lock, flags);
> + raw_spin_unlock_irqrestore(&x->wait.lock, flags);
> return true;
> }
> EXPORT_SYMBOL(completion_done);
>
