After calling device_initialize(), the reference count of the device is set to 1. If device_add() fails or register_queue_kobjects() fails, the function returns without calling put_device() to release the initial reference, causing a memory leak of the device structure. Similarly, in netdev_unregister_kobject(), after calling device_del(), there is no call to put_device() to release the initial reference, leading to a memory leak. Add put_device() in the error paths of netdev_register_kobject() and after device_del() in netdev_unregister_kobject() to properly release the device references.
Found by code review.
Cc: stable@vger.kernel.org Fixes: a1b3f594dc5f ("net: Expose all network devices in a namespaces in sysfs") Signed-off-by: Ma Ke make24@iscas.ac.cn --- net/core/net-sysfs.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index ca878525ad7c..d3895f26a0c8 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -2327,6 +2327,7 @@ void netdev_unregister_kobject(struct net_device *ndev) pm_runtime_set_memalloc_noio(dev, false);
device_del(dev); + put_device(dev); }
/* Create sysfs entries for network device. */ @@ -2357,7 +2358,7 @@ int netdev_register_kobject(struct net_device *ndev)
error = device_add(dev); if (error) - return error; + goto out_put_device;
error = register_queue_kobjects(ndev); if (error) { @@ -2367,6 +2368,10 @@ int netdev_register_kobject(struct net_device *ndev)
pm_runtime_set_memalloc_noio(dev, true);
+ return 0; + +out_put_device: + put_device(dev); return error; }
syzbot ci has tested the following series
[v1] net: Fix error handling in netdev_register_kobject https://lore.kernel.org/all/20251107080117.15099-1-make24@iscas.ac.cn * [PATCH] net: Fix error handling in netdev_register_kobject
and found the following issue: KASAN: slab-use-after-free Read in netdev_run_todo
Full report is available here: https://ci.syzbot.org/series/29bd058e-ea85-48e2-9bb9-ff9c0214f12e
***
KASAN: slab-use-after-free Read in netdev_run_todo
tree: net-next URL: https://kernel.googlesource.com/pub/scm/linux/kernel/git/netdev/net-next.git base: 8a25a2e34157d882032112e4194ccdfb29c499e8 arch: amd64 compiler: Debian clang version 20.1.8 (++20250708063551+0c9f909b7976-1~exp1~20250708183702.136), Debian LLD 20.1.8 config: https://ci.syzbot.org/builds/2f985280-f9ef-47c8-b4ec-80d44968222b/config
================================================================== BUG: KASAN: slab-use-after-free in kobject_put+0x2ab/0x480 Read of size 1 at addr ffff888113992714 by task kworker/u8:0/12
CPU: 0 UID: 0 PID: 12 Comm: kworker/u8:0 Not tainted syzkaller #0 PREEMPT(full) Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.2-debian-1.16.2-1 04/01/2014 Workqueue: netns cleanup_net Call Trace: <TASK> dump_stack_lvl+0x189/0x250 print_report+0xca/0x240 kasan_report+0x118/0x150 kobject_put+0x2ab/0x480 netdev_run_todo+0xd2e/0xea0 ops_undo_list+0x3e1/0x990 cleanup_net+0x4d8/0x820 process_scheduled_works+0xae1/0x17b0 worker_thread+0x8a0/0xda0 kthread+0x711/0x8a0 ret_from_fork+0x4bc/0x870 ret_from_fork_asm+0x1a/0x30 </TASK>
Allocated by task 5810: kasan_save_track+0x3e/0x80 __kasan_kmalloc+0x93/0xb0 __kvmalloc_node_noprof+0x5cd/0x910 alloc_netdev_mqs+0xa6/0x11b0 ip6gre_init_net+0xb5/0x3c0 ops_init+0x35c/0x5c0 setup_net+0xfe/0x320 copy_net_ns+0x34e/0x4e0 create_new_namespaces+0x3f3/0x720 unshare_nsproxy_namespaces+0x11c/0x170 ksys_unshare+0x4c8/0x8c0 __x64_sys_unshare+0x38/0x50 do_syscall_64+0xfa/0xfa0 entry_SYSCALL_64_after_hwframe+0x77/0x7f
Freed by task 12: kasan_save_track+0x3e/0x80 __kasan_save_free_info+0x46/0x50 __kasan_slab_free+0x5c/0x80 kfree+0x19a/0x6d0 device_release+0x9c/0x1c0 kobject_put+0x22b/0x480 netdev_run_todo+0xd0c/0xea0 ops_undo_list+0x3e1/0x990 cleanup_net+0x4d8/0x820 process_scheduled_works+0xae1/0x17b0 worker_thread+0x8a0/0xda0 kthread+0x711/0x8a0 ret_from_fork+0x4bc/0x870 ret_from_fork_asm+0x1a/0x30
The buggy address belongs to the object at ffff888113992000 which belongs to the cache kmalloc-cg-4k of size 4096 The buggy address is located 1812 bytes inside of freed 4096-byte region [ffff888113992000, ffff888113993000)
The buggy address belongs to the physical page: page: refcount:0 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x113990 head: order:3 mapcount:0 entire_mapcount:0 nr_pages_mapped:0 pincount:0 flags: 0x17ff00000000040(head|node=0|zone=2|lastcpupid=0x7ff) page_type: f5(slab) raw: 017ff00000000040 ffff88810004b500 dead000000000122 0000000000000000 raw: 0000000000000000 0000000000040004 00000000f5000000 0000000000000000 head: 017ff00000000040 ffff88810004b500 dead000000000122 0000000000000000 head: 0000000000000000 0000000000040004 00000000f5000000 0000000000000000 head: 017ff00000000003 ffffea00044e6401 00000000ffffffff 00000000ffffffff head: ffffffffffffffff 0000000000000000 00000000ffffffff 0000000000000008 page dumped because: kasan: bad access detected page_owner tracks the page as allocated page last allocated via order 3, migratetype Unmovable, gfp_mask 0xd20c0(__GFP_IO|__GFP_FS|__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP|__GFP_NOMEMALLOC), pid 5810, tgid 5810 (syz-executor), ts 65154411256, free_ts 60842795356 post_alloc_hook+0x240/0x2a0 get_page_from_freelist+0x2365/0x2440 __alloc_frozen_pages_noprof+0x181/0x370 alloc_pages_mpol+0x232/0x4a0 allocate_slab+0x96/0x350 ___slab_alloc+0xe94/0x18a0 __slab_alloc+0x65/0x100 __kmalloc_noprof+0x471/0x7f0 __register_sysctl_table+0x72/0x1340 __addrconf_sysctl_register+0x328/0x4c0 addrconf_sysctl_register+0x168/0x1c0 ipv6_add_dev+0xd46/0x1370 addrconf_notify+0x794/0x1010 notifier_call_chain+0x1b6/0x3e0 register_netdevice+0x1608/0x1ae0 register_netdev+0x40/0x60 page last free pid 15 tgid 15 stack trace: __free_frozen_pages+0xbc4/0xd30 __folio_put+0x21b/0x2c0 skb_release_data+0x49a/0x7c0 napi_consume_skb+0x158/0x1e0 skb_defer_free_flush+0x18f/0x250 net_rx_action+0x804/0xe50 handle_softirqs+0x286/0x870 run_ksoftirqd+0x9b/0x100 smpboot_thread_fn+0x542/0xa60 kthread+0x711/0x8a0 ret_from_fork+0x4bc/0x870 ret_from_fork_asm+0x1a/0x30
Memory state around the buggy address: ffff888113992600: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ffff888113992680: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
ffff888113992700: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
^ ffff888113992780: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ffff888113992800: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ==================================================================
***
If these findings have caused you to resend the series or submit a separate fix, please add the following tag to your commit message: Tested-by: syzbot@syzkaller.appspotmail.com
--- This report is generated by a bot. It may contain errors. syzbot ci engineers can be reached at syzkaller@googlegroups.com.
On Fri, Nov 7, 2025 at 12:01 AM Ma Ke make24@iscas.ac.cn wrote:
After calling device_initialize(), the reference count of the device is set to 1. If device_add() fails or register_queue_kobjects() fails, the function returns without calling put_device() to release the initial reference, causing a memory leak of the device structure. Similarly, in netdev_unregister_kobject(), after calling device_del(), there is no call to put_device() to release the initial reference, leading to a memory leak. Add put_device() in the error paths of netdev_register_kobject() and after device_del() in netdev_unregister_kobject() to properly release the device references.
Found by code review.
Cc: stable@vger.kernel.org Fixes: a1b3f594dc5f ("net: Expose all network devices in a namespaces in sysfs") Signed-off-by: Ma Ke make24@iscas.ac.cn
net/core/net-sysfs.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index ca878525ad7c..d3895f26a0c8 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -2327,6 +2327,7 @@ void netdev_unregister_kobject(struct net_device *ndev) pm_runtime_set_memalloc_noio(dev, false);
device_del(dev);
put_device(dev);
Please take a look at free_netdev()
}
/* Create sysfs entries for network device. */ @@ -2357,7 +2358,7 @@ int netdev_register_kobject(struct net_device *ndev)
error = device_add(dev); if (error)
return error;
goto out_put_device; error = register_queue_kobjects(ndev); if (error) {@@ -2367,6 +2368,10 @@ int netdev_register_kobject(struct net_device *ndev)
pm_runtime_set_memalloc_noio(dev, true);
return 0;+out_put_device:
put_device(dev); return error;
This seems bogus.
Was your report based on AI or some tooling ?
You would think that syzbot would have found an issue a long time ago...
Hello,
kernel test robot noticed "BUG_kmalloc-#k:Object_corrupt" on:
commit: 6e383382971f4573e29f0e85c6e5667c8f5f46d6 ("[PATCH] net: Fix error handling in netdev_register_kobject") url: https://github.com/intel-lab-lkp/linux/commits/Ma-Ke/net-Fix-error-handling-... base: https://git.kernel.org/cgit/linux/kernel/git/davem/net-next.git 6fc33710cd6c55397e606eeb544bdf56ee87aae5 patch link: https://lore.kernel.org/all/20251107080117.15099-1-make24@iscas.ac.cn/ patch subject: [PATCH] net: Fix error handling in netdev_register_kobject
in testcase: boot
config: i386-randconfig-014-20251110 compiler: gcc-14 test machine: qemu-system-i386 -enable-kvm -cpu SandyBridge -smp 2 -m 4G
(please refer to attached dmesg/kmsg for entire log/backtrace)
If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags | Reported-by: kernel test robot oliver.sang@intel.com | Closes: https://lore.kernel.org/oe-lkp/202511121347.b3f6c2a3-lkp@intel.com
[ OK ] Started /etc/rc.local Compatibility. LKP: ttyS0: 250: skip deploy intel ucode as no ucode is specified [ 15.199659] rc.local[250]: LKP: stdout: 250: skip deploy intel ucode as no ucode is specified [ 15.508990][ T1] [Poison overwritten] 0xf5e12370-0xf5e12370 @offset=9072. First byte 0x6a instead of 0x6b [ 15.511968][ T1] ============================================================================= [ 15.514232][ T1] BUG kmalloc-2k (Tainted: G S ): Object corrupt [ 15.516308][ T1] ----------------------------------------------------------------------------- [ 15.516308][ T1] [ 15.519187][ T1] Allocated in alloc_netdev_mqs+0x7d/0x470 age=517 cpu=1 pid=191 [ 15.521218][ T1] __slab_alloc+0x53/0x80 [ 15.522561][ T1] __kvmalloc_node_noprof (kbuild/src/consumer/mm/slub.c:4846 kbuild/src/consumer/mm/slub.c:5268 kbuild/src/consumer/mm/slub.c:5641 kbuild/src/consumer/mm/slub.c:7100)
The kernel config and materials to reproduce are available at: https://download.01.org/0day-ci/archive/20251112/202511121347.b3f6c2a3-lkp@i...
linux-stable-mirror@lists.linaro.org
-
Eric Dumazet -
kernel test robot -
Ma Ke -
syzbot ci