Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1416136

Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep

From Hannes Frederic Sowa <hannes@stressinduktion.org>
Newsgroups linux.kernel
Subject Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep
Date 2016-06-07 14:50 +0200
Message-ID <rHoCe-2KD-15@gated-at.bofh.it> (permalink)
References (6 earlier) <rFXuq-3Iu-27@gated-at.bofh.it> <rFXE6-3Ng-41@gated-at.bofh.it> <rGy0W-2a4-1@gated-at.bofh.it> <rH6vE-7U4-43@gated-at.bofh.it> <rHjsR-82q-7@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw

On 07.06.2016 09:15, Peter Zijlstra wrote:
>>
>> diff --git a/Documentation/memory-barriers.txt b/Documentation/memory-barriers.txt
>> index 147ae8ec836f..a4d0a99de04d 100644
>> --- a/Documentation/memory-barriers.txt
>> +++ b/Documentation/memory-barriers.txt
>> @@ -806,6 +806,41 @@ out-guess your code.  More generally, although READ_ONCE() does force
>>  the compiler to actually emit code for a given load, it does not force
>>  the compiler to use the results.
>>  
>> +In addition, control dependencies apply only to the then-clause and
>> +else-clause of the if-statement in question.  In particular, it does
>> +not necessarily apply to code following the if-statement:
>> +
>> +	q = READ_ONCE(a);
>> +	if (q) {
>> +		WRITE_ONCE(b, p);
>> +	} else {
>> +		WRITE_ONCE(b, r);
>> +	}
>> +	WRITE_ONCE(c, 1);  /* BUG: No ordering against the read from "a". */
>> +
>> +It is tempting to argue that there in fact is ordering because the
>> +compiler cannot reorder volatile accesses and also cannot reorder
>> +the writes to "b" with the condition.  Unfortunately for this line
>> +of reasoning, the compiler might compile the two writes to "b" as
>> +conditional-move instructions, as in this fanciful pseudo-assembly
>> +language:

I wonder if we already guarantee by kernel compiler settings that this
behavior is not allowed by at least gcc.

We unconditionally set --param allow-store-data-races=0 which should
actually prevent gcc from generating such conditional stores.

Am I seeing this correct here?

Thanks,
Hannes

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep Vineet Gupta <Vineet.Gupta1@synopsys.com> - 2016-06-03 11:20 +0200
  Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep Peter Zijlstra <peterz@infradead.org> - 2016-06-03 11:40 +0200
    Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-06-03 14:10 +0200
      Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep Peter Zijlstra <peterz@infradead.org> - 2016-06-03 14:30 +0200
        Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep Peter Zijlstra <peterz@infradead.org> - 2016-06-03 14:30 +0200
          Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-06-03 15:40 +0200
        Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-06-03 15:40 +0200
          Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep Will Deacon <will.deacon@arm.com> - 2016-06-03 15:50 +0200
            Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-06-05 06:40 +0200
              Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-06-06 19:30 +0200
                Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep Peter Zijlstra <peterz@infradead.org> - 2016-06-07 09:20 +0200
                Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep Hannes Frederic Sowa <hannes@stressinduktion.org> - 2016-06-07 14:50 +0200
                Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-06-07 15:10 +0200
                Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep Hannes Frederic Sowa <hannes@stressinduktion.org> - 2016-06-07 17:00 +0200
                Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-06-07 17:30 +0200
                Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep Peter Zijlstra <peterz@infradead.org> - 2016-06-07 19:50 +0200
                Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-06-07 20:50 +0200
                Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep Will Deacon <will.deacon@arm.com> - 2016-06-07 20:10 +0200
                Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-06-07 20:50 +0200
                Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-06-07 21:00 +0200
                Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep Hannes Frederic Sowa <hannes@stressinduktion.org> - 2016-06-07 20:40 +0200

csiph-web