RE: [PATCH PR95855]A missing ifcvt optimization to generate fcsel

2020-06-30 Thread yangyang (ET)
> On Tue, Jun 30, 2020 at 1:31 PM yangyang (ET) > wrote: > > > > > On Tue, Jun 30, 2020 at 4:29 AM yangyang (ET) > > > > > > wrote: > > > > > > > > Hi, > > > > > > > > > > Hi, > > > > > > > > > > > > This is a simple fix for pr95855. > > > > > > > > > > > > With this fix,

Re: [PATCH PR95855]A missing ifcvt optimization to generate fcsel

2020-06-30 Thread Richard Biener via Gcc-patches
On Tue, Jun 30, 2020 at 1:31 PM yangyang (ET) wrote: > > > On Tue, Jun 30, 2020 at 4:29 AM yangyang (ET) > > wrote: > > > > > > Hi, > > > > > > > > Hi, > > > > > > > > > > This is a simple fix for pr95855. > > > > > > > > > > With this fix, pass_split_paths can recognize the > > > > >

RE: [PATCH PR95855]A missing ifcvt optimization to generate fcsel

2020-06-30 Thread yangyang (ET)
> On Tue, Jun 30, 2020 at 4:29 AM yangyang (ET) > wrote: > > > > Hi, > > > > > > Hi, > > > > > > > > This is a simple fix for pr95855. > > > > > > > > With this fix, pass_split_paths can recognize the > > > > if-conversion > > > opportunity of the testcase and doesn't duplicate the

Re: [PATCH PR95855]A missing ifcvt optimization to generate fcsel

2020-06-30 Thread Richard Biener via Gcc-patches
On Tue, Jun 30, 2020 at 4:29 AM yangyang (ET) wrote: > > Hi, > > > > Hi, > > > > > > This is a simple fix for pr95855. > > > > > > With this fix, pass_split_paths can recognize the if-conversion > > opportunity of the testcase and doesn't duplicate the corresponding block. > > > > > >

RE: [PATCH PR95855]A missing ifcvt optimization to generate fcsel

2020-06-29 Thread yangyang (ET)
Hi, > > Hi, > > > > This is a simple fix for pr95855. > > > > With this fix, pass_split_paths can recognize the if-conversion > opportunity of the testcase and doesn't duplicate the corresponding block. > > > > Added one testcase for this. Bootstrap and tested on both aarch64 and >

Re: [PATCH PR95855]A missing ifcvt optimization to generate fcsel

2020-06-29 Thread Richard Biener via Gcc-patches
On Sun, Jun 28, 2020 at 2:32 PM yangyang (ET) wrote: > > Hi, > > This is a simple fix for pr95855. > > With this fix, pass_split_paths can recognize the if-conversion > opportunity of the testcase and doesn't duplicate the corresponding block. > > Added one testcase for this.