[PATCH] D29866: [PDSE] Add PDSE.

Daniel Berlin via llvm-commits llvm-commits at lists.llvm.org
Tue Jun 20 10:42:39 PDT 2017


On Tue, Jun 20, 2017 at 4:42 AM, Tobias Grosser <tobias at grosser.es> wrote:

> On Tue, Jun 20, 2017, at 01:03 PM, Daniel Berlin via llvm-commits wrote:
> > On Tue, Jun 20, 2017 at 12:26 AM, Tobias Grosser <tobias at grosser.es>
> > wrote:
> >
> > > I think the PDT brokenness is something all of us want to see resolved.
> > > I just had an interesting discussion with Hal how currently the PDT is
> > > not usable to test the validity of liferange metadata.
> > >
> > > I think it would be great if we collect the use case we have for
> > > dominator trees to clearly see where they are "broken" today and which
> > > solution could address this brokenness.
> >
> >
> > I'm not sure why you put broken in quotes?
>
> Because we can only reason about broken, if we know what is correct. In
> the presence of unreachables and infinite loops this is not a
> straightforward extension. I know you claim it is trivial, but that's
> not clear.
>

I'm going to assert, again, it is, because there is no way to make a valid
dominator tree otherwise.
But like i said, you are welcome to try!



>
> Anyway, let's wait for the verifier.
>
> > > With Jakub Kuderski pushing
> > > non-trivial improvements to dominance relations in LLVM we seem to have
> > > a unique chance to get this fixed and well documented in a way that
> > > either satisfies all of us -- or at least clearly documents the
> > > tradeoffs that have been taken in the final result.
> > >
> >
> > Jakub has a verifier he's about to post that will verify the structural
> > properties of the trees are valid.
> > If you can make your version/tradeoffs verify, have at it.
>
> I think a verifier is a good step, especially if formalizes the
> invariants you would like.
>

Again, please be very careful with the language you are using.
These are not the invariants "i would like".
These are the two invariants of "what makes a dominator tree".  If you are
going to claim their is contention on this point, this is pointless.
It is not me who came up with them, they have been the same since the 70's.

As I mentioned, you are welcome to do anything to the post-dominator tree
you like that maintains them.

I think you are going to discover, quite quickly, there is only one way to
maintain these invariants.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20170620/228d9cb5/attachment.html>


More information about the llvm-commits mailing list