Revision 580e4273d7a883ececfefa692c1f96bdbacb99b5 authored by Cong Wang on 02 October 2020, 19:13:34 UTC, committed by David S. Miller on 04 October 2020, 21:53:06 UTC
Although we take RTNL on dump path, it is possible to
skip RTNL on insertion path. So the following race condition
is possible:
rtnl_lock() // no rtnl lock
mutex_lock(&idrinfo->lock);
// insert ERR_PTR(-EBUSY)
mutex_unlock(&idrinfo->lock);
tc_dump_action()
rtnl_unlock()
So we have to skip those temporary -EBUSY entries on dump path
too.
Reported-and-tested-by: syzbot+b47bc4f247856fb4d9e1@syzkaller.appspotmail.com
Fixes: 0fedc63fadf0 ("net_sched: commit action insertions together")
Cc: Vlad Buslov <vladbu@mellanox.com>
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
1 parent 9a9e774
| File | Mode | Size |
|---|---|---|
| Kconfig | -rw-r--r-- | 4.2 KB |
| Makefile | -rw-r--r-- | 952 bytes |
| badrange.c | -rw-r--r-- | 7.7 KB |
| blk.c | -rw-r--r-- | 8.3 KB |
| btt.c | -rw-r--r-- | 43.1 KB |
| btt.h | -rw-r--r-- | 7.3 KB |
| btt_devs.c | -rw-r--r-- | 8.9 KB |
| bus.c | -rw-r--r-- | 31.3 KB |
| claim.c | -rw-r--r-- | 7.6 KB |
| core.c | -rw-r--r-- | 13.8 KB |
| dax_devs.c | -rw-r--r-- | 2.8 KB |
| dimm.c | -rw-r--r-- | 3.1 KB |
| dimm_devs.c | -rw-r--r-- | 24.7 KB |
| e820.c | -rw-r--r-- | 1.8 KB |
| label.c | -rw-r--r-- | 35.2 KB |
| label.h | -rw-r--r-- | 4.1 KB |
| namespace_devs.c | -rw-r--r-- | 67.1 KB |
| nd-core.h | -rw-r--r-- | 7.7 KB |
| nd.h | -rw-r--r-- | 11.1 KB |
| nd_virtio.c | -rw-r--r-- | 3.6 KB |
| of_pmem.c | -rw-r--r-- | 2.6 KB |
| pfn.h | -rw-r--r-- | 1.5 KB |
| pfn_devs.c | -rw-r--r-- | 21.8 KB |
| pmem.c | -rw-r--r-- | 16.8 KB |
| pmem.h | -rw-r--r-- | 1.1 KB |
| region.c | -rw-r--r-- | 4.1 KB |
| region_devs.c | -rw-r--r-- | 33.1 KB |
| security.c | -rw-r--r-- | 13.9 KB |
| virtio_pmem.c | -rw-r--r-- | 3.0 KB |
| virtio_pmem.h | -rw-r--r-- | 1.3 KB |
Computing file changes ...
