linux-kernel - Re: [PATCH v2] rcu/nocb: Add Safe checks for access offloaded rdp

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: <aBuK29q6nuhTgHIq@localhost.localdomain>
Date: Wed, 7 May 2025 18:31:23 +0200
From: Frederic Weisbecker <frederic@...nel.org>
To: Joel Fernandes <joelagnelf@...dia.com>
Cc: Zqiang <qiang.zhang1211@...il.com>, paulmck@...nel.org,
	neeraj.upadhyay@...nel.org, joel@...lfernandes.org,
	urezki@...il.com, boqun.feng@...il.com, rcu@...r.kernel.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] rcu/nocb: Add Safe checks for access offloaded rdp

Le Wed, May 07, 2025 at 12:06:29PM -0400, Joel Fernandes a écrit :
> 
> 
> On 5/7/2025 7:26 AM, Zqiang wrote:
> > For built with CONFIG_PROVE_RCU=y and CONFIG_PREEMPT_RT=y kernels,
> > Disable BH does not change the SOFTIRQ corresponding bits in
> > preempt_count(), but change current->softirq_disable_cnt, this
> > resulted in the following splat:
> > 
> > WARNING: suspicious RCU usage
> > kernel/rcu/tree_plugin.h:36 Unsafe read of RCU_NOCB offloaded state!
> > stack backtrace:
> > CPU: 0 UID: 0 PID: 22 Comm: rcuc/0
> > Call Trace:
> > [    0.407907]  <TASK>
> > [    0.407910]  dump_stack_lvl+0xbb/0xd0
> > [    0.407917]  dump_stack+0x14/0x20
> > [    0.407920]  lockdep_rcu_suspicious+0x133/0x210
> > [    0.407932]  rcu_rdp_is_offloaded+0x1c3/0x270
> > [    0.407939]  rcu_core+0x471/0x900
> > [    0.407942]  ? lockdep_hardirqs_on+0xd5/0x160
> > [    0.407954]  rcu_cpu_kthread+0x25f/0x870
> > [    0.407959]  ? __pfx_rcu_cpu_kthread+0x10/0x10
> > [    0.407966]  smpboot_thread_fn+0x34c/0xa50
> > [    0.407970]  ? trace_preempt_on+0x54/0x120
> > [    0.407977]  ? __pfx_smpboot_thread_fn+0x10/0x10
> > [    0.407982]  kthread+0x40e/0x840
> > [    0.407990]  ? __pfx_kthread+0x10/0x10
> > [    0.407994]  ? rt_spin_unlock+0x4e/0xb0
> > [    0.407997]  ? rt_spin_unlock+0x4e/0xb0
> > [    0.408000]  ? __pfx_kthread+0x10/0x10
> > [    0.408006]  ? __pfx_kthread+0x10/0x10
> > [    0.408011]  ret_from_fork+0x40/0x70
> > [    0.408013]  ? __pfx_kthread+0x10/0x10
> > [    0.408018]  ret_from_fork_asm+0x1a/0x30
> > [    0.408042]  </TASK>
> > 
> > Currently, triggering an rdp offloaded state change need the
> > corresponding rdp's CPU goes offline, and at this time the rcuc
> > kthreads has already in parking state. this means the corresponding
> > rcuc kthreads can safely read offloaded state of rdp while it's
> > corresponding cpu is online.
> > 
> > This commit therefore add softirq_count() check for
> > Preempt-RT kernels.
> > 
> > Suggested-by: Joel Fernandes <joelagnelf@...dia.com>
> > Signed-off-by: Zqiang <qiang.zhang1211@...il.com>
> > ---
> >  kernel/rcu/tree_plugin.h | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
> > index 003e549f6514..a91b2322a0cd 100644
> > --- a/kernel/rcu/tree_plugin.h
> > +++ b/kernel/rcu/tree_plugin.h
> > @@ -29,7 +29,7 @@ static bool rcu_rdp_is_offloaded(struct rcu_data *rdp)
> >  		  (IS_ENABLED(CONFIG_HOTPLUG_CPU) && lockdep_is_cpus_held()) ||
> >  		  lockdep_is_held(&rdp->nocb_lock) ||
> >  		  lockdep_is_held(&rcu_state.nocb_mutex) ||
> > -		  (!(IS_ENABLED(CONFIG_PREEMPT_COUNT) && preemptible()) &&
> > +		  ((!(IS_ENABLED(CONFIG_PREEMPT_COUNT) && preemptible()) || softirq_count()) &&
> >  		   rdp == this_cpu_ptr(&rcu_data)) ||
> This looks good to me. Frederic told me he'll further review and give final
> green signal. Then I'll pull this particular one.
> 
> One thing I was wondering -- it would be really nice if preemptible() itself
> checked for softirq_count() by default. Or adding something like a
> really_preemptible() which checks for both CONFIG_PREEMPT_COUNT and
> softirq_count() along with preemptible().  I feel like this always comes back to
> bite us in different ways, and not knowing atomicity complicates various code paths.
> 
> Maybe a summer holidays project? ;)

I thought about that too but I think this is semantically incorrect.
In PREEMPT_RT, softirqs are actually preemptible.

Thanks.

> 
>  - Joel
> 

-- 
Frederic Weisbecker
SUSE Labs