linux-kernel - Re: [bisected] system hang after boot

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]
Date:   Mon, 27 Nov 2017 13:10:31 +0000
From:   Will Deacon <will.deacon@....com>
To:     Martin Schwidefsky <schwidefsky@...ibm.com>
Cc:     Peter Zijlstra <peterz@...radead.org>,
        Sebastian Ott <sebott@...ux.vnet.ibm.com>,
        Ingo Molnar <mingo@...nel.org>,
        Heiko Carstens <heiko.carstens@...ibm.com>,
        linux-kernel@...r.kernel.org
Subject: Re: [bisected] system hang after boot

On Mon, Nov 27, 2017 at 02:00:28PM +0100, Martin Schwidefsky wrote:
> On Mon, 27 Nov 2017 12:54:56 +0000
> Will Deacon <will.deacon@....com> wrote:
> > On Mon, Nov 27, 2017 at 01:49:18PM +0100, Martin Schwidefsky wrote:
> > > On Mon, 27 Nov 2017 11:49:48 +0000
> > > Will Deacon <will.deacon@....com> wrote:  
> > > > On Wed, Nov 22, 2017 at 09:22:17PM +0100, Peter Zijlstra wrote:  
> > > > > On Wed, Nov 22, 2017 at 06:26:59PM +0000, Will Deacon wrote:
> > > > >     
> > > > > > Now, I can't see what the break_lock is doing here other than causing
> > > > > > problems. Is there a good reason for it, or can you just try removing it
> > > > > > altogether? Patch below.    
> > > > > 
> > > > > The main use is spin_is_contended(), which in turn ends up used in
> > > > > __cond_resched_lock() through spin_needbreak().
> > > > > 
> > > > > This allows better lock wait times for PREEMPT kernels on platforms
> > > > > where the lock implementation itself cannot provide 'contended' state.
> > > > > 
> > > > > In that capacity the write-write race shouldn't be a problem though.    
> > > > 
> > > > I'm not sure why it isn't a problem: given that the break_lock variable
> > > > can read as 1 for a lock that is no longer contended and 0 for a lock that
> > > > is currently contended, then the __cond_resched_lock is likely to see a
> > > > value of 0 (i.e. spin_needbreak always return false) more often than no
> > > > since it's checked by the lock holder.  
> > > 
> > > Grepping for 'break_lock' the two locking blueprints are the only places
> > > where the field is written to. Unless I am blind, the associated unlock
> > > functions do *not* reset 'break_lock'.
> > > 
> > > Without the raw_##op##_can_lock(lock) check the first of the blueprints
> > > now looks like this:
> > > 
> > > void __lockfunc __raw_##op##_lock(locktype##_t *lock)                   \
> > > {                                                                       \
> > >         for (;;) {                                                      \
> > >                 preempt_disable();                                      \
> > >                 if (likely(do_raw_##op##_trylock(lock)))                \
> > >                         break;                                          \
> > >                 preempt_enable();                                       \
> > >                                                                         \
> > >                 if (!(lock)->break_lock)                                \
> > >                         (lock)->break_lock = 1;                         \
> > >                 while ((lock)->break_lock)                              \
> > >                         arch_##op##_relax(&lock->raw_lock);             \
> > >         }                                                               \
> > >         (lock)->break_lock = 0;                                         \
> > > }                                                                       \
> > > 
> > > All it takes to create an endless loop is two CPUs, the first acquired the
> > > lock and the second tries to get the lock. After the unsuccessful trylock
> > > of the second CPU, the first CPU releases the lock and never tries to take
> > > it again. The second CPU will be stuck in an endless loop.  
> > 
> > Yes, it basically relies on the lock holder never winning that race.
> > However, Peter's use-case just needs the lock-holder to be able to detect
> > contention (which is always best-effort anyway), so I think we can make that
> > "work" by removing the while loop above (see my subsequent diff sent to
> > Sebastian).
> 
> Well, what race? The lock hold just has to hold the lock while another CPU
> tries to get it. There is no particular bad timing involved, just a little
> bit of contention is enough.

Yes, you're right. I keep forgetting that break_lock isn't cleared on
unlock.

> And yes, I think removing the while loop on break_lock will work.
> 
> > It's still questionable, because on a machine with store-buffers you really
> > want to order writes to break_lock against something else, but it might
> > happen to fall out depending on the details of the trylock() implementation.
> 
> Even more, if the compiler "proves" that nobody writes to break_lock it can
> convert that to "while (1)" loop.

break_lock should be annotated (at least) with READ_ONCE/WRITE_ONCE, which
should prevent that from happening.

> > > I guess my best course of action is to remove GENERIC_LOCKBREAK from
> > > arch/s390/Kconfig to avoid this construct altogether. Let us see what
> > > breaks if I do that ..  
> > 
> > We could just consider ripping out GENERIC_LOCKBREAK entirely, but I was
> > hoping we could get a simpler fix in for now.
> 
> I would opt for removing it entirely.

I'll cook a patch series, with the first patch just removing the while loop
and subsequent patches removing the stuff altogether.

Will