[llvm-commits] [llvm] r172534 - /llvm/trunk/test/Transforms/LoopStrengthReduce/post-inc-icmpzero.ll

[Renato Golin]

On 15 January 2013 17:44, Andrew Trick <atrick at apple.com> wrote:
> The test case specifies a triple. I agree it would be nice to know how this test was affected by the host. I'd also like to know if this broke after some recent checkin. This may just be a symptom of a real issue with cross-compilation.

Yes, sorry, you are right, the triple is there.

I agree it's important to know how each host affects the test,
however, the test is specifically asking for a sequence of
instructions that is relevant to the problem it's trying to test, and
that patter is still matched.

I don't think we gain much by examining each difference in tests
between hosts, since LLVM has no intention to produce the exact same
IR on all hosts, even with the same triple.

> Manish, can you at least post the difference between IR output from ARM vs. x86?

I'm generating it again and will attach here later, if he doesn't beat
me to it. ;)

cheers,
--renato