netdev - Re: [oops] with FIB

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <18960.5308.297655.51837@robur.slu.se>
Date:	Sun, 17 May 2009 15:44:28 +0200
From:	Robert Olsson <robert@...ur.slu.se>
To:	paulmck@...ux.vnet.ibm.com
Cc:	Robert Olsson <Robert.Olsson@...a.slu.se>,
	Stephen Hemminger <shemminger@...tta.com>,
	andrei.popa@...eo.ro, NetDEV list <netdev@...r.kernel.org>
Subject: Re: [oops] with FIB_TRIE


Paul E. McKenney writes:

 > >  Need to recap this... at least for myself insert, delete and flush is 
 > >  serialised wrt RTNL I guess this is still OK. We oops in lookup but 
 > >  we should have preempt_disable as we're in a RCU read section. Also we 
 > >  know quagga is constantly running delete...
 > 
 > RCU read-side critical sections are not preempt_disable() if you have
 > built with CONFIG_PREEMPT_RCU.

I've just fixed a setup so we can verify the oops easily in the lab. 

# CONFIG_PREEMPT_RCU is not set
# CONFIG_PREEMPT_RCU_TRACE is not set
# CONFIG_PREEMPT_NONE is not set
CONFIG_PREEMPT_VOLUNTARY=y
# CONFIG_PREEMPT is not set

Any suggestion where to start? 
I'll guess we'll need to instrument a bit and try to understand what happens. 
softirq can be run by ksoftirqd

CONFIG_PREEMPT=y
ftp://robur.slu.se/pub/Linux/net-development/trie-test/OOPS

Cheers
					--ro

Stephen. rcu_read_lock_bh in trie_loop didn't make any difference
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html