linux-kernel - Re: [PATCH 2/2] lockdep: fix 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] [thread-next>] [day] [month] [year] [list]

Message-Id: <20080312.230843.114525346.davem@davemloft.net>
Date:	Wed, 12 Mar 2008 23:08:43 -0700 (PDT)
From:	David Miller <davem@...emloft.net>
To:	a.p.zijlstra@...llo.nl
Cc:	linux-kernel@...r.kernel.org, mingo@...e.hu, ego@...ibm.com,
	hugh@...itas.com, dada1@...mosbay.com
Subject: Re: [PATCH 2/2] lockdep: fix fib_hash softirq inversion

From: Peter Zijlstra <a.p.zijlstra@...llo.nl>
Date: Wed, 12 Mar 2008 13:09:22 +0100

> =========================================================
> [ INFO: possible irq lock inversion dependency detected ]
> 2.6.25-rc4-sched-devel.git #56
> ---------------------------------------------------------
> swapper/0 just changed the state of lock:
>  (&rt_hash_locks[i]){-+..}, at: [<ffffffff804721ac>] rt_intern_hash+0x8c/0x3b0
> but this lock took another, soft-read-irq-unsafe lock in the past:
>  (fib_hash_lock){-.-?}
> 
> and interrupts could create inverse lock ordering between them.

I tried to figure out what lockdep doesn't like here.

Could you show me the specific code path that could cause
the lock conflict?

Adding BH disabling to fib_hash_lock will add non-trivial
costs to these code paths, so I'd like to avoid this if
possible.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/