Re: [PATCH, rs6000] Fix vec_construct vectorization cost to be somewhat more accurate

2016-08-11 Thread Segher Boessenkool
On Wed, Aug 10, 2016 at 11:52:06AM -0500, Bill Schmidt wrote: > Bootstrapped and tested on powerpc64le-unknown-linux-gnu with no > regressions. Ok for trunk? Okay. Thanks, Segher > 2016-08-10 Bill Schmidt > > * config/rs6000/rs6000.c

Re: [PATCH, rs6000] Fix vec_construct vectorization cost to be somewhat more accurate

2016-08-10 Thread Bill Schmidt
Sorry for the long delay on getting back to this. I took a look at the suggested test cases with the cost model available, and did some SPEC testing to validate the model. I found that it is still important to model the 4xfloat case separately to account for conversion from 64-bit to 32-bit in

Re: [PATCH, rs6000] Fix vec_construct vectorization cost to be somewhat more accurate

2016-07-18 Thread Richard Biener
On Mon, Jul 18, 2016 at 1:56 PM, Segher Boessenkool wrote: > Hi Bill, > > On Fri, Jul 15, 2016 at 08:55:08AM -0500, Bill Schmidt wrote: >> This patch is a follow-up to Richard's patch of >> https://gcc.gnu.org/ml/gcc-patches/2016-06/msg00584.html. The cost of a >>

Re: [PATCH, rs6000] Fix vec_construct vectorization cost to be somewhat more accurate

2016-07-18 Thread Segher Boessenkool
Hi Bill, On Fri, Jul 15, 2016 at 08:55:08AM -0500, Bill Schmidt wrote: > This patch is a follow-up to Richard's patch of > https://gcc.gnu.org/ml/gcc-patches/2016-06/msg00584.html. The cost of a > vec_construct (initialization of an N-way vector by N scalars) is too low, > which can cause

Re: [PATCH, rs6000] Fix vec_construct vectorization cost to be somewhat more accurate

2016-07-18 Thread Richard Biener
On Fri, Jul 15, 2016 at 3:55 PM, Bill Schmidt wrote: > Hi, > > This patch is a follow-up to Richard's patch of > https://gcc.gnu.org/ml/gcc-patches/2016-06/msg00584.html. The cost of a > vec_construct (initialization of an N-way vector by N scalars) is too low, >

[PATCH, rs6000] Fix vec_construct vectorization cost to be somewhat more accurate

2016-07-15 Thread Bill Schmidt
Hi, This patch is a follow-up to Richard's patch of https://gcc.gnu.org/ml/gcc-patches/2016-06/msg00584.html. The cost of a vec_construct (initialization of an N-way vector by N scalars) is too low, which can cause too-aggressive vectorization in particular for N=8 or higher. Richard changed