[PATCH] Ignore lifetime intrinsics in use list for MemCpyOptimizer.

[Chandler Carruth]

On Mon, Sep 1, 2014 at 2:06 PM, Hal Finkel <hfinkel at anl.gov> wrote:

> > Doh, sorry. Before commit I spent a while thinking about where
> > lifetime intrinsics could be placed such that ignoring them would
> > make the transformation invalid (ie., you start relying on a
> > lifetime.end'd value where previously you didn't). I didn't actually
> > review the implementation.
> >
>
> So did I, and moreover, I did in theory review the implementation. I
> think, however, that I saw what I expected to see and not what was actually
> there :(
>

Not a big deal at all. =]

Mostly, this is a good case study in the utility of negative tests to check
that what we expect to see and think we see is in fact what is there.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140901/f5d8ed5b/attachment.html>