Re: [PATCH net-next 1/2] mac80211_hwsim: Make hwsim_netgroup IDA

Am 2=2E M=C3=A4rz 2018 12:37:25 MEZ schrieb Kirill Tkhai <ktkhai@virtuozzo=
=2Ecom>:

destroy_radio() may be executed in parallel with everything above you
wrote,
doesn't it? There may be several network namespaces, and
destroy_radio()
queued from one net namespace may race with mac80211_hwsim_new_radio()
or hwsim_del_radio_nl() for another net namespace=2E I don't see, how
netlink
locking can act on synchronization with a work=2E This is what I mention=

=2E

I see, you are right=2E Nonetheless, this value is pretty uncritical, sinc=
e the user (the netlink dump) only checks whether it changes within a dump =
and even if there would be race conditions, e=2Eg=2E some generations would=
 be skipped caused by parallel writing, it would also set the dump interrup=
ted flag, and the user space program knows, if it needs exact results, it n=
eeds to dump again=2E I'm unsure about things like caching of this variable=
=2E Maybe it needs a volatile flag to work always as expected=2E

Unfortunately, currently the code triggers a dump interrupted also when th=
e interfaces of the current namespace didn't change, but I think that is ac=
ceptable=2E Otherwise we need a per namespace generation and I think all th=
is happens really rare and it's not worth the effort=2E

Thanks,
Kirill

--=20
M=2ESc=2E Benjamin Beichler

Universit=C3=A4t Rostock, Fakult=C3=A4t f=C3=BCr Informatik und Elektrotec=
hnik
Institut f=C3=BCr Angewandte Mikroelektronik und Datentechnik

University of Rostock, Department of CS and EE
Institute of Applied Microelectronics and CE

Richard-Wagner-Stra=C3=9Fe 31
18119 Rostock
Deutschland/Germany

phone: +49 (0) 381 498 - 7278
email: Benjamin=2EBeichler@uni-rostock=2Ede
www: http://www=2Eimd=2Euni-rostock=2Ede/