[LLVMdev] Vectorization: Next Steps
tobias at grosser.es
Tue Feb 7 08:17:51 PST 2012
On 02/07/2012 05:12 PM, David A. Greene wrote:
> Preston Briggs<preston.briggs at gmail.com> writes:
>> On Mon, Feb 6, 2012 at 12:13 PM, Sebastian Pop<spop at codeaurora.org> wrote:
>>> [many things, but I'm only going to focus on one of them]
>>> Would you consider using Polly http://polly.grosser.es to avoid
>>> writing this code?
>> My impression is that Polly (and polyhedral analysis generally)
>> doesn't do want I want. But I'm happy to talk about it 'cause I might
>> be wrong.
> I think you are right. However, Polly might be useful as one layer in a
> dependence tester. I haven't thought too deeply about it but polyhedral
> analysis can do some pretty cool things.
> I'm really not clear on what something like CLooG would do with the
> linear recurrence example. But the analysis side of Polly might be
> quite useful.
CLooG does not do anything to it. It is just a layer to lower a high
level abstraction back to imperative control flow. The question her is,
if the linear recurrence example and similar code can be easily
optimized within the high level polyhedral abstraction without having to
rewrite on the LLVM-IR layer. If you provide the expected output, I may
give this a try.
More information about the llvm-dev