Re: [PATCH 05/11] locking/ww_mutex: Add waiters in stamp order

[Chris Wilson]

On Mon, Nov 28, 2016 at 01:20:06PM +0100, Nicolai Hähnle wrote:
> From: Nicolai Hähnle <Nicolai.Haehnle@xxxxxxx>
> 
> Add regular waiters in stamp order. Keep adding waiters that have no
> context in FIFO order and take care not to starve them.
> 
> While adding our task as a waiter, back off if we detect that there is a
> waiter with a lower stamp in front of us.
> 
> Make sure to call lock_contended even when we back off early.

I'm hitting
[   86.202749] WARNING: CPU: 1 PID: 813 at ./include/linux/ww_mutex.h:292 stress_inorder_work+0x436/0x4b5 [test_ww_mutex]
[   86.202885] DEBUG_LOCKS_WARN_ON(!ctx->contending_lock)

which if I understand correctly is due to

> +static inline int __sched
> +__ww_mutex_add_waiter(struct mutex_waiter *waiter,
> +		      struct mutex *lock,
> +		      struct ww_acquire_ctx *ww_ctx)
> +{
> +	if (ww_ctx) {
> +		struct mutex_waiter *cur;
> +
> +		/*
> +		 * Add the waiter before the first waiter with a higher stamp.
> +		 * Waiters without a context are skipped to avoid starving
> +		 * them.
> +		 */
> +		list_for_each_entry(cur, &lock->wait_list, list) {
> +			if (!cur->ww_ctx)
> +				continue;
> +
> +			if (__ww_mutex_stamp_after(ww_ctx, cur->ww_ctx)) {
> +				/* Back off immediately if necessary. */
> +				if (ww_ctx->acquired > 0)
> +					return -EDEADLK;

not setting ww_ctx->contending_lock here.

> +
> +				continue;
> +			}
> +
> +			list_add_tail(&waiter->list, &cur->list);
> +			return 0;
> +		}
> +	}
> +
> +	list_add_tail(&waiter->list, &lock->wait_list);
> +	return 0;
> +}

-- 
Chris Wilson, Intel Open Source Technology Centre