On Wed, Jul 22, 2020 at 05:01:54PM +0200, Peter Zijlstra wrote:
> While the traditional irq_work relies on the ability to self-IPI, it
> makes sense to provide an unconditional irq_work_queue_remote()
> interface.
>
> This can be used to replace the plagued smp_call_function_single_async().
>
> Signed-off-by: Peter Zijlstra (Intel) <pet...@infradead.org>
> ---
> include/linux/irq_work.h | 17 ++++--
> kernel/irq_work.c | 129
> ++++++++++++++++++++++++++++-------------------
> kernel/rcu/tree.c | 6 +-
> 3 files changed, 95 insertions(+), 57 deletions(-)
[ . . . ]
> --- a/kernel/irq_work.c
> +++ b/kernel/irq_work.c
[ . . . ]
> @@ -196,17 +237,5 @@ void irq_work_tick(void)
> irq_work_run_list(this_cpu_ptr(&lazy_list));
> }
>
> -/*
> - * Synchronize against the irq_work @entry, ensures the entry is not
> - * currently in use.
> - */
> -void irq_work_sync(struct irq_work *work)
> -{
> - lockdep_assert_irqs_enabled();
> -
> - while (irq_work_is_busy(work))
> - cpu_relax();
> -}
> -EXPORT_SYMBOL_GPL(irq_work_sync);
> -
> #endif /* CONFIG_IRQ_WORK */
> +
FYI, "git am" complains about this blank line.
Thanx, Paul
