Re: [PATCH][1/n] Fix PR81303

2017-07-21 Thread Richard Biener
On Fri, 21 Jul 2017, Bin.Cheng wrote: > On Fri, Jul 21, 2017 at 8:12 AM, Richard Biener wrote: > > > > The following is sth I noticed when looking at a way to fix PR81303. > > We happily compute a runtime cost model threshold that executes the > > vectorized variant even

Re: [PATCH][1/n] Fix PR81303

2017-07-21 Thread Bin.Cheng
On Fri, Jul 21, 2017 at 8:12 AM, Richard Biener wrote: > > The following is sth I noticed when looking at a way to fix PR81303. > We happily compute a runtime cost model threshold that executes the > vectorized variant even though no vector iteration takes place due > to the

[PATCH][1/n] Fix PR81303

2017-07-21 Thread Richard Biener
The following is sth I noticed when looking at a way to fix PR81303. We happily compute a runtime cost model threshold that executes the vectorized variant even though no vector iteration takes place due to the number of prologue/epilogue iterations. The following fixes that -- note that if we