[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Sat, 5 May 2018 20:51:45 +0200
From: Andrew Lunn <andrew@...n.ch>
To: Kees Cook <keescook@...omium.org>
Cc: Salvatore Mesoraca <s.mesoraca16@...il.com>,
Florian Fainelli <f.fainelli@...il.com>,
Vivien Didelot <vivien.didelot@...oirfairelinux.com>,
LKML <linux-kernel@...r.kernel.org>,
Kernel Hardening <kernel-hardening@...ts.openwall.com>,
Network Development <netdev@...r.kernel.org>,
"David S. Miller" <davem@...emloft.net>
Subject: Re: [PATCH] net: dsa: drop some VLAs in switch.c
> > You could make the bitmap part of the dsa_switch structure. This is
> > allocated by dsa_switch_alloc() and is passed the number of ports.
> > Doing the allocation there means you don't need to worry about it
> > failing in dsa_switch_mdb_add() or dsa_switch_vlan_add().
>
> Are dsa_switch_mdb_add() and dsa_switch_vlan_add() guaranteed to be
> single-threaded?
Yes, that is the interesting question here.... against each other, or
themselves?
They are called from a notifier chain. It is the same notifier chain
for both dsa_switch_mdb_add() and dsa_switch_vlan_add().
notifier_call_chain() itself appears to not provide any guarantees
about the same handler being called in parallel.
It is dsa_port_notify() which is calling the notifier_call_chain().
This is being called by both dsa_port_vlan_add() and
dsa_port_mdb_add() in dsa_slave_port_obj_add(). This is a switchdev
op. switchdev_port_obj_add_now() does have ASSERT_RTNL(); So that
should serialize everything.
Andrew
Powered by blists - more mailing lists