Re: [PATCH 11/12] Generic semaphore implementation

[Harvey Harrison <harvey.harrison@gmail.com>]

> +/*
> + * Because this function is inlined, the 'state' parameter will be constant,
> + * and thus optimised away by the compiler.
> + */

If so, is unlikely() still needed?

> +static inline int __sched __down_common(struct semaphore *sem, long state)
> +{
> +	int result = 0;
> +	struct task_struct *task = current;
> +	struct semaphore_waiter waiter;
> +
> +	list_add_tail(&waiter.list, &sem->wait_list);
> +	waiter.task = task;
> +	waiter.up = 0;
> +
> +	for (;;) {
> +		if (unlikely((state == TASK_INTERRUPTIBLE &&
> +					signal_pending(task)) ||
> +			     (state == TASK_KILLABLE &&
> +					fatal_signal_pending(task))))
> +			goto interrupted;

		if (state == TASK_INTERRUPTIBLE && signal_pending(task))
			goto interrupted;
		if (state == TASK_KILLABLE && fatal_signal_pending(task))
			goto interrupted;

Other than that small bit, looks great, any thoughts to getting rid of
all the likely/unlikely as well?  If they're no longer performance
critical, it would be nice to have it that little bit cleaner.


Harvey