From 4d92b95ff2f95f13df9bad0b5a25a9f60e72758d Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 4 Dec 2021 20:21:57 -0800 Subject: net: add net device refcount tracker infrastructure net device are refcounted. Over the years we had numerous bugs caused by imbalanced dev_hold() and dev_put() calls. The general idea is to be able to precisely pair each decrement with a corresponding prior increment. Both share a cookie, basically a pointer to private data storing stack traces. This patch adds dev_hold_track() and dev_put_track(). To use these helpers, each data structure owning a refcount should also use a "netdevice_tracker" to pair the hold and put. netdevice_tracker dev_tracker; ... dev_hold_track(dev, &dev_tracker, GFP_ATOMIC); ... dev_put_track(dev, &dev_tracker); Whenever a leak happens, we will get precise stack traces of the point dev_hold_track() happened, at device dismantle phase. We will also get a stack trace if too many dev_put_track() for the same netdevice_tracker are attempted. This is guarded by CONFIG_NET_DEV_REFCNT_TRACKER option. Signed-off-by: Eric Dumazet Signed-off-by: Jakub Kicinski --- net/Kconfig.debug | 10 ++++++++++ net/core/dev.c | 3 +++ 2 files changed, 13 insertions(+) create mode 100644 net/Kconfig.debug (limited to 'net') diff --git a/net/Kconfig.debug b/net/Kconfig.debug new file mode 100644 index 000000000000..fb5c70e01cb3 --- /dev/null +++ b/net/Kconfig.debug @@ -0,0 +1,10 @@ +# SPDX-License-Identifier: GPL-2.0-only + +config NET_DEV_REFCNT_TRACKER + bool "Enable net device refcount tracking" + depends on DEBUG_KERNEL && STACKTRACE_SUPPORT + select REF_TRACKER + default n + help + Enable debugging feature to track device references. + This adds memory and cpu costs. diff --git a/net/core/dev.c b/net/core/dev.c index aba8acc1238c..1740d6cfe86b 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -9864,6 +9864,7 @@ static void netdev_wait_allrefs(struct net_device *dev) netdev_unregister_timeout_secs * HZ)) { pr_emerg("unregister_netdevice: waiting for %s to become free. Usage count = %d\n", dev->name, refcnt); + ref_tracker_dir_print(&dev->refcnt_tracker, 10); warning_time = jiffies; } } @@ -10154,6 +10155,7 @@ struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name, dev = PTR_ALIGN(p, NETDEV_ALIGN); dev->padded = (char *)dev - (char *)p; + ref_tracker_dir_init(&dev->refcnt_tracker, 128); #ifdef CONFIG_PCPU_DEV_REFCNT dev->pcpu_refcnt = alloc_percpu(int); if (!dev->pcpu_refcnt) @@ -10270,6 +10272,7 @@ void free_netdev(struct net_device *dev) list_for_each_entry_safe(p, n, &dev->napi_list, dev_list) netif_napi_del(p); + ref_tracker_dir_exit(&dev->refcnt_tracker); #ifdef CONFIG_PCPU_DEV_REFCNT free_percpu(dev->pcpu_refcnt); dev->pcpu_refcnt = NULL; -- cgit