[llvm-dev] Possible soundness issue with available_externally (split from "RFC: Add guard intrinsics")

[Duncan P. N. Exon Smith via llvm-dev]

> On 2016-Feb-24, at 19:46, Sanjoy Das <sanjoy at playingwithpointers.com> wrote:
> 
> On Wed, Feb 24, 2016 at 7:38 PM, Chandler Carruth <chandlerc at google.com> wrote:
>> On Wed, Feb 24, 2016 at 7:34 PM Duncan P. N. Exon Smith
>> <dexonsmith at apple.com> wrote:
>>> 
>>> 
>>>> On 2016-Feb-24, at 19:17, Chandler Carruth <chandlerc at google.com> wrote:
>>>> 
>>>> On Wed, Feb 24, 2016 at 7:10 PM Sanjoy Das via llvm-dev
>>>> <llvm-dev at lists.llvm.org> wrote:
>>>> On Wed, Feb 24, 2016 at 6:51 PM, Duncan P. N. Exon Smith
>>>> <dexonsmith at apple.com> wrote:
>>>>>> If we do not inline @foo(), and instead re-link the call site in
>>>>>> @main
>>>>>> to some non-optimized copy (or differently optimized copy) of @foo,
>>>>>> then it is possible for the program to have the behavior {print("Y");
>>>>>> print ("X")}, which was disallowed in the earlier program.
>>>>>> 
>>>>>> In other words, opt refined the semantics of @foo() (i.e. reduced the
>>>>>> set of behaviors it may have) in ways that would make later
>>>>>> optimizations invalid if we de-refine the implementation of @foo().
>>>>> 
>>>>> I'm probably missing something obvious here.  How could the result of
>>>>> `%t0 != %t1` be different at optimization time in one file than from
>>>>> runtime in the "real" implementation?  Doesn't this make the CSE
>>>>> invalid?
>>>> 
>>>> `%t0` and `%t1` are "allowed" to "always be the same", i.e. an
>>>> implementation of @foo that always feeds in the same
>>>> value for `%t0` and `%t1` is a valid implementation (which is why the
>>>> CSE was valid); but it is not the *only* valid implementation.  If I
>>>> don't CSE the two load instructions (also a valid thing to do), and
>>>> this is a second thread writing to `%par`, then the two values loaded
>>>> can be different, and you could end up printing `"X"` in `@foo`.
>>>> 
>>>> Did that make sense?
>>> 
>>> Yes.  To be sure I understand the scope: this is only a problem for
>>> atomics, correct?  (Because multi-threaded behaviour with other globals
>>> is UB?)
>>> 
>>>>> Does linkonce_odr linkage have the same problem?
>>>>> - If so, do you want to change it too?
>>>>> - Else, why not?
>>>> 
>>>> Going by the specification in the LangRef, I'd say it depends on how
>>>> you define "definitive".  If you're allowed to replace the body of a
>>>> function with a differently optimized body, then the above problem
>>>> exists.
>>>> 
>>>> I believe that is the case, and I strongly believe the problem you
>>>> outline exists for linkonce_odr exactly as it does for available_externally.
>>>> 
>>>> Which is what makes this scary: every C++ inline function today can
>>>> trigger this.
>>> 
>>> Every C/C++ inline or template function.  But only the ones that use
>>> atomics, right?
>> 
>> 
>> Well, with *this* example...
> 
> Atomic are one source of non-determinism that compilers can reason
> about.  I don't know if the following snippet is well defined or not,
> but you could have similar issues with
> 
> 
>  void foo() {
>    int *p = malloc(sizeof(int));
>    if (*p < 10) print("X");
>  }
> 
> or (again, I don't know if this is actually well defined)
> 
>  void foo() {
>    int t;  // it is probably reasonable to fold compares with
> ptrtoint(alloca) to undef
>    if ((intptr_t)(&t) < 10) print("X");
>  }
> 

The first one at least is UB, but as Richard pointed out the scope
is certainly broader than atomics (it's not even just well-defined
non-deterministism).

I'm kind of terrified by the implications.

> -- Sanjoy
> 
>> 
>>> 
>>> 
>>> Not that I'm sure that will end up being a helpful distinction.
>> 
>> 
>> Right. See Richard's comment. I think that sums up the real issue here. =/