| commit | 350d6591c9670767e1002919dc866c83dddfc806 | |
| author | Eric Dumazet <edumazet@google.com> | |
| Thu, 16 May 2019 15:09:57 +0000 (16 08:09 -0700) | ||
| committer | Greg Kroah-Hartman <gregkh@linuxfoundation.org> | |
| Sat, 25 May 2019 16:26:43 +0000 (25 18:26 +0200) | ||
| tree | d2d6857fc9cadbe82541915f35265f5b1e2eb1ab | treesnapshot (tar.gz zip) |
| parent | a5f56b52c878585b12b8bc37f737dcce4a660c64 | commitdiff |
net: avoid weird emergency message
[ Upstream commit d7c04b05c9ca14c55309eb139430283a45c4c25f ]
When host is under high stress, it is very possible thread
running netdev_wait_allrefs() returns from msleep(250)
10 seconds late.
This leads to these messages in the syslog :
[...] unregister_netdevice: waiting for syz_tun to become free. Usage count = 0
If the device refcount is zero, the wait is over.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit d7c04b05c9ca14c55309eb139430283a45c4c25f ]
When host is under high stress, it is very possible thread
running netdev_wait_allrefs() returns from msleep(250)
10 seconds late.
This leads to these messages in the syslog :
[...] unregister_netdevice: waiting for syz_tun to become free. Usage count = 0
If the device refcount is zero, the wait is over.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
| net/core/dev.c | diffblobblamehistory |