[LLVMdev] How to broaden the SLP vectorizer's search

[Roel Jordans]

On 08/08/14 17:27, Frank Winter wrote:
> I changed the max. recursion depth to 36, and tried then 1000 (from the
> original value of 12) and it did not improve SLP's optimization
> capabilities on my input function. For example, the attached function is
> (by design) perfectly vectorizable into 4-packed single precision SIMD
> load/add/store. The SLP vectorizer just does nothing with it.
>
> I ran
>
> opt -default-data-layout="e-m:e-i64:64-f80:128-n8:16:32:64-S128"
> -basicaa -slp-vectorizer -S < mod_vec_p_vec.ll
>

The data layout that you're specifying doesn't have a vector type 
included?  Maybe adding one will help

http://llvm.org/docs/LangRef.html#data-layout

Cheers,
  Roel

> with RecursionMaxDepth = 12, 36, and 1000.
>
> Thanks,
> Frank
>
>
> On 08/07/2014 12:57 PM, Renato Golin wrote:
>> On 7 August 2014 17:33, Chad Rosier <mcrosier at codeaurora.org> wrote:
>>> You might consider filing a bug (llvm.org/bugs) requesting a flag, but I
>>> don't know if the code owners want to expose such a flag.
>> I'm not sure that's a good idea as a raw access to that limit, as
>> there are no guarantees that it'll stay the same. But maybe a flag
>> turning some "aggressive" behaviour from SLP that would then change
>> that threshold (and maybe some others) would be a good flag to have, I
>> think.
>>
>> This could maybe be a way to deprecate the BB vectorizer faster than
>> we would otherwise. But that would depend on all missing BB features
>> to be implemented in SLP.
>>
>> An item in bugzilla seems appropriate.
>>
>> cheers,
>> --renato
>
>
>
>
> _______________________________________________
> LLVM Developers mailing list
> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>