gro_cells allocates a per-CPU struct gro_cell with per-CPU queues (cell->napi_skbs), per-CPU local_lock_t (cell->bh_lock), and per-CPU napi_struct (cell->napi). On !CONFIG_PREEMPT_RT, local_lock_t only provides lockdep annotation (and is a complete no-op when CONFIG_DEBUG_LOCK_ALLOC is disabled), relying on per-CPU locality and disabled BH context for mutual exclusion; it does not provide cross-CPU synchronization. When threaded NAPI is enabled on a device backed by gro_cells (e.g. gre0, vxlan, geneve) via sysfs or Netlink, unbound kernel threads are created for each per-CPU NAPI. When gro_cells_receive() on CPU A enqueues a packet to cell_A and calls napi_schedule(&cell_A->napi) while holding cell_A->bh_lock, the woken kthread can run concurrently on remote CPU B. When CPU B runs gro_cell_poll(&cell_A->napi) and calls __local_lock_nested_bh(&cell_A->bh_lock), lockdep detects that the lock is already held by CPU A's task and triggers a warning: WARNING: CPU: 1 PID: 377 at net/core/gro_cells.c:66 gro_cell_poll DEBUG_LOCKS_WARN_ON(l->owner) CPU: 1 UID: 0 PID: 377 Comm: napi/gre0-0 Not tainted 7.3.0-rc2 Call Trace: __local_lock_nested_bh include/linux/local_lock_internal.h:92 gro_cell_poll+0x5aa/0x6c0 net/core/gro_cells.c:66 napi_threaded_poll+0x39b/0x600 net/core/dev.c:7048 kthread+0x71e/0x870 kernel/kthread.c:464 ret_from_fork+0x5d/0x90 arch/x86/kernel/process.c:167 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:264 On !CONFIG_PREEMPT_RT kernels without lockdep, concurrent execution of __skb_queue_tail() on CPU A and __skb_dequeue() on CPU B without cross-CPU locks leads to silent queue corruption. Virtual per-CPU NAPIs cannot support unbound threaded NAPI. Just as gro_cells sets NAPI_STATE_NO_BUSY_POLL because virtual NAPIs cannot support busy polling, introduce NAPI_STATE_NO_THREADED to indicate that threaded mode is not supported. Set NAPI_STATE_NO_THREADED in gro_cells_init(), and fail early with -EOPNOTSUPP when trying to enable threaded napi on gro_cells netdevs. Fixes: 25718fdcbdd2 ("net: gro_cells: Use nested-BH locking for gro_cell") Reported-by: syzbot+f0661448aa9511ce744a@syzkaller.appspotmail.com Closes: https://lore.kernel.org/netdev/6aad6d7b.0c43d342.320d00.0001.GAE@google.com Reviewed-by: Kuniyuki Iwashima Signed-off-by: Naman Gulati --- include/linux/netdevice.h | 2 ++ net/core/gro_cells.c | 1 + net/core/net-sysfs.c | 4 +++- net/core/netdev-genl.c | 3 +++ 4 files changed, 9 insertions(+), 1 deletion(-) diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 87cafc932e9e..951734bc8b36 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -433,6 +433,7 @@ enum { NAPI_STATE_SCHED_THREADED, /* Napi is currently scheduled in threaded mode */ NAPI_STATE_HAS_NOTIFIER, /* Napi has an IRQ notifier */ NAPI_STATE_THREADED_BUSY_POLL, /* The threaded NAPI poller will busy poll */ + NAPI_STATE_NO_THREADED, /* Threaded mode is not supported */ }; enum { @@ -448,6 +449,7 @@ enum { NAPIF_STATE_SCHED_THREADED = BIT(NAPI_STATE_SCHED_THREADED), NAPIF_STATE_HAS_NOTIFIER = BIT(NAPI_STATE_HAS_NOTIFIER), NAPIF_STATE_THREADED_BUSY_POLL = BIT(NAPI_STATE_THREADED_BUSY_POLL), + NAPIF_STATE_NO_THREADED = BIT(NAPI_STATE_NO_THREADED), }; enum gro_result { diff --git a/net/core/gro_cells.c b/net/core/gro_cells.c index d8c0a2867120..9f8884d56f59 100644 --- a/net/core/gro_cells.c +++ b/net/core/gro_cells.c @@ -92,6 +92,7 @@ int gro_cells_init(struct gro_cells *gcells, struct net_device *dev) local_lock_init(&cell->bh_lock); set_bit(NAPI_STATE_NO_BUSY_POLL, &cell->napi.state); + set_bit(NAPI_STATE_NO_THREADED, &cell->napi.state); netif_napi_add(dev, &cell->napi, gro_cell_poll); napi_enable(&cell->napi); diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index 352173df7578..296fdc007862 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -743,9 +743,11 @@ static ssize_t threaded_show(struct device *dev, static int modify_napi_threaded(struct net_device *dev, unsigned long val) { + struct napi_struct *napi; int ret; - if (list_empty(&dev->napi_list)) + napi = list_first_entry_or_null(&dev->napi_list, typeof(*napi), dev_list); + if (!napi || test_bit(NAPI_STATE_NO_THREADED, &napi->state)) return -EOPNOTSUPP; if (val != 0 && val != 1) diff --git a/net/core/netdev-genl.c b/net/core/netdev-genl.c index cb18db681640..661f87e2f015 100644 --- a/net/core/netdev-genl.c +++ b/net/core/netdev-genl.c @@ -334,6 +334,9 @@ netdev_nl_napi_set_config(struct napi_struct *napi, struct genl_info *info) if (info->attrs[NETDEV_A_NAPI_THREADED]) { int ret; + if (test_bit(NAPI_STATE_NO_THREADED, &napi->state)) + return -EOPNOTSUPP; + threaded = nla_get_uint(info->attrs[NETDEV_A_NAPI_THREADED]); ret = napi_set_threaded(napi, threaded); if (ret) -- 2.55.0.1082.g2b9226bbc0-goog