linux-kernel - Re: [PATCH v2] bit_spinlock: introduce smp_cond_load

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: <20181106110026.GJ22431@hirez.programming.kicks-ass.net>
Date:   Tue, 6 Nov 2018 12:00:26 +0100
From:   Peter Zijlstra <peterz@...radead.org>
To:     Gao Xiang <gaoxiang25@...wei.com>
Cc:     Will Deacon <will.deacon@....com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Philippe Ombredanne <pombredanne@...b.com>,
        Kate Stewart <kstewart@...uxfoundation.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        linux-kernel@...r.kernel.org, Miao Xie <miaoxie@...wei.com>,
        Chao Yu <chao@...nel.org>
Subject: Re: [PATCH v2] bit_spinlock: introduce smp_cond_load_relaxed

On Tue, Nov 06, 2018 at 06:22:31PM +0800, Gao Xiang wrote:
> Hi Peter,
> 
> On 2018/11/6 17:06, Peter Zijlstra wrote:
> > On Mon, Nov 05, 2018 at 10:49:21PM +0000, Will Deacon wrote:
> >> diff --git a/include/asm-generic/bitops/lock.h b/include/asm-generic/bitops/lock.h
> >> index 3ae021368f48..9de8d3544630 100644
> >> --- a/include/asm-generic/bitops/lock.h
> >> +++ b/include/asm-generic/bitops/lock.h
> >> @@ -6,6 +6,15 @@
> >>  #include <linux/compiler.h>
> >>  #include <asm/barrier.h>
> >>  
> >> +static inline void spin_until_bit_unlock(unsigned int nr,
> >> +					 volatile unsigned long *p)
> >> +{
> >> +	unsigned long mask = BIT_MASK(bitnum);
> >> +
> >> +	p += BIT_WORD(nr);
> >> +	smp_cond_load_relaxed(p, VAL & mask);
> >> +}
> >> +
> >>  /**
> >>   * test_and_set_bit_lock - Set a bit and return its old value, for lock
> >>   * @nr: Bit to set
> >> diff --git a/include/linux/bit_spinlock.h b/include/linux/bit_spinlock.h
> >> index bbc4730a6505..d711c62e718c 100644
> >> --- a/include/linux/bit_spinlock.h
> >> +++ b/include/linux/bit_spinlock.h
> >> @@ -26,9 +26,7 @@ static inline void bit_spin_lock(int bitnum, unsigned long *addr)
> >>  #if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK)
> >>  	while (unlikely(test_and_set_bit_lock(bitnum, addr))) {
> >>  		preempt_enable();
> >> -		do {
> >> -			cpu_relax();
> >> -		} while (test_bit(bitnum, addr));
> >> +		spin_until_bit_unlock(bitnum, addr);
> >>  		preempt_disable();
> >>  	}
> >>  #endif
> > 
> > Yes, that's much better. Ideally though, we'd get rid of bit spinlocks
> > that have significant enough contention for this to matter.
> 
> OK, I will send v3 to fix like the above.

That's not answering the full question though. What bit spinlocks did
you find where this matters? And can't we convert them to proper
spinlocks instead?