Revision ff1fa2081d173b01cebe2fbf0a2d0f1cee9ce4b5 authored by Jakub Kicinski on 29 June 2022, 18:19:10 UTC, committed by Jakub Kicinski on 30 June 2022, 18:34:10 UTC
Eric reports that syzbot made short work out of my speculative
fix. Indeed when queue gets detached its tfile->tun remains,
so we would try to stop NAPI twice with a detach(), close()
sequence.

Alternative fix would be to move tun_napi_disable() to
tun_detach_all() and let the NAPI run after the queue
has been detached.

Fixes: a8fc8cb5692a ("net: tun: stop NAPI when detaching queues")
Reported-by: syzbot <syzkaller@googlegroups.com>
Reported-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20220629181911.372047-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
1 parent 9c5de24
History
File Mode Size
Kconfig -rw-r--r-- 529 bytes
Makefile -rw-r--r-- 378 bytes
ast.c -rw-r--r-- 8.2 KB
ast.h -rw-r--r-- 1.1 KB
config.c -rw-r--r-- 23.2 KB
config.h -rw-r--r-- 1.4 KB
debug_fs.c -rw-r--r-- 21.5 KB
dir.c -rw-r--r-- 7.3 KB
dir.h -rw-r--r-- 889 bytes
dlm_internal.h -rw-r--r-- 20.8 KB
lock.c -rw-r--r-- 162.2 KB
lock.h -rw-r--r-- 3.2 KB
lockspace.c -rw-r--r-- 21.0 KB
lockspace.h -rw-r--r-- 940 bytes
lowcomms.c -rw-r--r-- 47.2 KB
lowcomms.h -rw-r--r-- 1.9 KB
lvb_table.h -rw-r--r-- 541 bytes
main.c -rw-r--r-- 1.9 KB
member.c -rw-r--r-- 16.4 KB
member.h -rw-r--r-- 1.3 KB
memory.c -rw-r--r-- 3.1 KB
memory.h -rw-r--r-- 1.2 KB
midcomms.c -rw-r--r-- 37.9 KB
midcomms.h -rw-r--r-- 1.5 KB
netlink.c -rw-r--r-- 2.8 KB
plock.c -rw-r--r-- 11.1 KB
rcom.c -rw-r--r-- 17.0 KB
rcom.h -rw-r--r-- 987 bytes
recover.c -rw-r--r-- 22.8 KB
recover.h -rw-r--r-- 1.3 KB
recoverd.c -rw-r--r-- 8.1 KB
recoverd.h -rw-r--r-- 781 bytes
requestqueue.c -rw-r--r-- 5.0 KB
requestqueue.h -rw-r--r-- 736 bytes
user.c -rw-r--r-- 24.5 KB
user.h -rw-r--r-- 440 bytes
util.c -rw-r--r-- 1.6 KB
util.h -rw-r--r-- 545 bytes

back to top