Re: [PATCH AArch64]Penalize vector cost for large loops with too many vect insns.

> On Oct 17, 2016, at 12:26 PM, Bin.Cheng wrote: > > On Sat, Oct 15, 2016 at 3:07 AM, kugan > wrote: >> Hi Bin, >> >> On 15/10/16 00:15, Bin Cheng wrote: >>> >>> +/* Test for likely overcommitment of vector hardware resources. If a

Re: [PATCH AArch64]Penalize vector cost for large loops with too many vect insns.

On Sat, Oct 15, 2016 at 3:07 AM, kugan wrote: > Hi Bin, > > On 15/10/16 00:15, Bin Cheng wrote: >> >> +/* Test for likely overcommitment of vector hardware resources. If a >> + loop iteration is relatively large, and too large a percentage of >> +

Re: [PATCH AArch64]Penalize vector cost for large loops with too many vect insns.

On Sat, Oct 15, 2016 at 01:07:12PM +1100, kugan wrote: > Hi Bin, > > On 15/10/16 00:15, Bin Cheng wrote: > >+/* Test for likely overcommitment of vector hardware resources. If a > >+ loop iteration is relatively large, and too large a percentage of > >+ instructions in the loop are

Re: [PATCH AArch64]Penalize vector cost for large loops with too many vect insns.

Hi Bin, On 15/10/16 00:15, Bin Cheng wrote: +/* Test for likely overcommitment of vector hardware resources. If a + loop iteration is relatively large, and too large a percentage of + instructions in the loop are vectorized, the cost model may not + adequately reflect delays from

[PATCH AArch64]Penalize vector cost for large loops with too many vect insns.

Hi, It is suspected that for some large loops with too many vect_insns after vectorization, the performance will be regressed because of various reasons. Two possible reasons are: 1) Suggested by powerpc backend, large loops with too many vect_insns may jam vector unit's pipeline. 2)