Re: [tip:locking/urgent] compiler, atomics: Provide READ_ONCE_NOCHECK ()

[Dmitry Vyukov]

On Wed, Oct 14, 2015 at 6:16 PM, Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:
> On Wed, Oct 14, 2015 at 06:08:16PM +0200, Dmitry Vyukov wrote:
>> >> > So I add READ_ONCE_NOCHECK() for accesses for which the compiler cannot
>> >> > prove safe address for KASAN's benefit, but READ_ONCE() suffices for
>> >> > the data-race-detection logic in KTSAN, correct?
>> >>
>> >> KTSAN also needs READ_ONCE_NOCHECK() here. KTSAN will flag races
>> >> between get_wchan() and the thread accesses to own stack even more
>> >> aggressively than KASAN, because KTSAN won't like get_wchan() accesses
>> >> even to non-poisoned areas of other thread stack.
>> >
>> > So to keep KTSAN happy, any read from some other thread's stack requires
>> > READ_ONCE_NOCHECK()?  What if the access is via a locking primitive or
>> > read-modify-write atomic operation?
>> >
>> > This is of some interest in RCU, which implements synchronous grace
>> > periods using completions that are allocated on the calling task's stack
>> > and manipulated by RCU callbacks that are likely executing elsewhere.
>>
>>
>> KTSAN does not have any special logic for stacks. It just generally
>> flags pairs of accesses when (1) at least one access is not atomic,
>> (2) at least one access is a write and (3) these accesses are not
>> synchronized by means of other synchronization.
>
> But but but.. WRITE_ONCE/READ_ONCE _are_ atomic when on naturally
> aligned machine word sized thingies. We very much rely on that.
>
> And the wchan thing is very much that, its not some weird large object,
> its a single word, read with an explicit 'volatile' cast.
>
> This is good, and should not require more magic annotations.


Well, if another thread writes it byte-by-byte, it pretty much does
not matter how you read it.
Note that I said "at least one access is not atomic". If both are
atomic, then this is, of course, legal. And KTSAN considers
READ/WRITE_ONCE as atomic operations.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/