net: gro_cells: Use nested-BH locking for gro_cell

The gro_cell data structure is per-CPU variable and relies on disabled
BH for its locking. Without per-CPU locking in local_bh_disable() on
PREEMPT_RT this data structure requires explicit locking.

Add a local_lock_t to the data structure and use
local_lock_nested_bh() for locking. This change adds only lockdep
coverage and does not alter the functional behaviour for !PREEMPT_RT.

Reported-by: [e-mail]
Closes: https://lore.kernel.org/all/[e-mail]/
Fixes: 3253cb49cbad ("softirq: Allow to drop the softirq-BKL lock on PREEMPT_RT")
Signed-off-by: Sebastian Andrzej Siewior <[e-mail]>
Link: https://patch.msgid.link/[e-mail]
Signed-off-by: Jakub Kicinski <[e-mail]>

net/core/gro_cells.c

···
 struct gro_cell {
 	struct sk_buff_head	napi_skbs;
 	struct napi_struct	napi;
+	local_lock_t		bh_lock;
 };
 
 int gro_cells_receive(struct gro_cells *gcells, struct sk_buff *skb)
 {
 	struct net_device *dev = skb->dev;
+	bool have_bh_lock = false;
 	struct gro_cell *cell;
 	int res;
 
···
 		goto unlock;
 	}
 
+	local_lock_nested_bh(&gcells->cells->bh_lock);
+	have_bh_lock = true;
 	cell = this_cpu_ptr(gcells->cells);
 
 	if (skb_queue_len(&cell->napi_skbs) > READ_ONCE(net_hotdata.max_backlog)) {
···
 	__skb_queue_tail(&cell->napi_skbs, skb);
 	if (skb_queue_len(&cell->napi_skbs) == 1)
 		napi_schedule(&cell->napi);
+
+	if (have_bh_lock)
+		local_unlock_nested_bh(&gcells->cells->bh_lock);
 
 	res = NET_RX_SUCCESS;
 
···
 	struct sk_buff *skb;
 	int work_done = 0;
 
+	__local_lock_nested_bh(&cell->bh_lock);
 	while (work_done < budget) {
 		skb = __skb_dequeue(&cell->napi_skbs);
 		if (!skb)
···
 
 	if (work_done < budget)
 		napi_complete_done(napi, work_done);
+	__local_unlock_nested_bh(&cell->bh_lock);
 	return work_done;
 }
 
···
 		struct gro_cell *cell = per_cpu_ptr(gcells->cells, i);
 
 		__skb_queue_head_init(&cell->napi_skbs);
+		local_lock_init(&cell->bh_lock);
 
 		set_bit(NAPI_STATE_NO_BUSY_POLL, &cell->napi.state);