[llvm-dev] [RFC] Introducing the maynotprogress IR attribute

[James Y Knight via llvm-dev]

On Fri, Sep 11, 2020 at 12:42 PM Atmn Patel <atmndp at gmail.com> wrote:

> Hi Hal,
>
> On Thu, Sep 10, 2020 at 8:54 PM Hal Finkel <hfinkel at anl.gov> wrote:
> >
> > Hi, Atmn,
> >
> > Has anyone else expressed an opinion regarding the naming? We need to
> > clarify the semantics in C, it seems.
>
> No other names have come in yet, in total the names proposed so far (I
> think) are:
> - maynotprogress
> - maybenoprogress
> - might_not_progress
> - nfpg
> - no_fpg
> and the loop metadata has been pretty firmly established as
> llvm.loop.mustprogress. IMHO, I've warmed up to no_fpg (or even nfpg
> in a pinch if we want to save 33% of space and lose some readability)
> since we've made substantial progress in clarifying the exact
> definitions of progress in this context and I think it's a good idea
> to bake it into the attribute name.


I'd actually like to suggest that we invert the default for
functions. Rather than adding a "maynotprogress" function attribute,
instead add a "mustprogress" function attribute, which Clang will emit on
every function compiled in C++ mode. For two reasons:
1. Having both the attribute and the loop metadata be the same way around
makes it simpler to think about (rather than one being positive, and the
other being negated).
2. Given that the global progress-requirement seems to be pretty much
C++-specific, having this behavior be off by default, and opted into by C++
frontends makes sense.
3. Bonus: it makes choosing an attribute name easier: mustprogress, done.

I've also modified the clang patch [0] to only apply either of the

attributes for C functions when compiled with C11 or later so we can
> tightly adhere to both the C and C++ standards, and the other changes
> that need to be made will be forthcoming. Thanks again to James, that
> particular example was pretty cool, and I agree that it may be best to
> follow that interpretation.
>
> [0] https://reviews.llvm.org/D86841


You mean that you now apply maynotprogress to all functions in C, right?
But why only C11 and later? I think all versions of C should get the
maynotprogress function attribute? (Or, with the change I suggest above:
only C++ code should get the "mustprogress" function attribute.)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20200911/3255b481/attachment.html>