[PATCH][LoopStrengthReduce] Don't trim a formula that uses a subset of required registers

[Chandler Carruth]

On Mon, Apr 28, 2014 at 5:12 PM, Adam Nemet <anemet at apple.com> wrote:

> OK, so as I understand this is the plan:
>
> * file a PR on the original LSR problem with the C test case
> * check in the fix again (without the test case), referencing the PR
> * keep the PR open at this point until we figure out a way to have an LLVM
> IR test case for this
>

I would file the PR with the C test case for "llc cannot reproduce LSR
behavior observed with clang" or some such. It should have nothing to do
with the particular bug fix, and the bug fix shouldn't reference the PR (we
don't ever reference PRs from source code).

Another way to look at this is that the PR should essentially be "make llc
able to test this, and commit the test case when that works".


> * remove clang from the substitutions in lit.cfg so that the next person
> won’t get confused like I got
>

Sure, but in a separate commit.


Despite all of this, I'm a little worried that this will get shelved in a
PR and no one will ever look at it. This kind of deficiency in our testing
infrastructure seems like a really scary and important thing, but its not
even something I know how to chase down (not being an expert on LSR).
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140428/b344fde1/attachment.html>