Re: [PATCH net] net: gro_cells: Use nested-BH locking for gro_cell
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: 2025-11-03 13:59:20
Also in:
linux-rt-devel
Subsystem:
networking [general], the rest · Maintainers:
"David S. Miller", Eric Dumazet, Jakub Kicinski, Paolo Abeni, Linus Torvalds
On 2025-11-03 04:36:45 [-0800], Eric Dumazet wrote:
Adding LOCKDEP annotations would be needed (like what we do in netdev_lockdep_set_classes()
You mean something like
diff --git a/include/net/gro_cells.h b/include/net/gro_cells.h
index 596688b67a2a8..1df6448701879 100644
--- a/include/net/gro_cells.h
+++ b/include/net/gro_cells.h@@ -10,6 +10,7 @@ struct gro_cell; struct gro_cells { struct gro_cell __percpu *cells; + struct lock_class_key cells_bh_class; }; int gro_cells_receive(struct gro_cells *gcells, struct sk_buff *skb);
diff --git a/net/core/gro_cells.c b/net/core/gro_cells.c
index fd57b845de333..1c98d32657e85 100644
--- a/net/core/gro_cells.c
+++ b/net/core/gro_cells.c@@ -88,6 +88,7 @@ int gro_cells_init(struct gro_cells *gcells, struct net_device *dev) __skb_queue_head_init(&cell->napi_skbs); local_lock_init(&cell->bh_lock); + lockdep_set_class(&cell->bh_lock, &gcells->cells_bh_class); set_bit(NAPI_STATE_NO_BUSY_POLL, &cell->napi.state);
Or I would try something like :
I'm fine with both. I can provide a patch body for either of the two. Sebastian