Re: [PATCH] On x86 allow if-conversion of more than one insn as long as there is at most one cmov (PR tree-optimization/79390)

2017-04-06 Thread Jakub Jelinek
On Thu, Apr 06, 2017 at 02:50:27PM +0200, Rainer Orth wrote: > 2017-04-06 Rainer Orth > > * gcc.target/i386/pr79390.c: Allow for cmovl.a. Please mention PR tree-optimization/79390 in the ChangeLog and commit message. Ok with that change. > # HG changeset

Re: [PATCH] On x86 allow if-conversion of more than one insn as long as there is at most one cmov (PR tree-optimization/79390)

2017-04-06 Thread Rainer Orth
Jeff Law writes: > On 04/01/2017 06:20 AM, Jakub Jelinek wrote: >> Hi! >> >> As discussed in the PR, in the following testcase we don't if-convert >> with the generic (and many other) tuning, because we default to >> --param max-rtl-if-conversion-insns=1 in most of the tunings.

Re: [PATCH] On x86 allow if-conversion of more than one insn as long as there is at most one cmov (PR tree-optimization/79390)

2017-04-04 Thread Jeff Law
On 04/01/2017 06:20 AM, Jakub Jelinek wrote: Hi! As discussed in the PR, in the following testcase we don't if-convert with the generic (and many other) tuning, because we default to --param max-rtl-if-conversion-insns=1 in most of the tunings. The problem we have is with multiple cmov

Re: [PATCH] On x86 allow if-conversion of more than one insn as long as there is at most one cmov (PR tree-optimization/79390)

2017-04-03 Thread Jakub Jelinek
On Sun, Apr 02, 2017 at 08:44:03PM +0200, Uros Bizjak wrote: > x86 part LGTM. > > Hopefully, this infrastructure will allow us to fix (or it already > fixes) PR 56309 [1]. I think only allows to. The target hook has access to the if_info structure which contains the original basic blocks, their

Re: [PATCH] On x86 allow if-conversion of more than one insn as long as there is at most one cmov (PR tree-optimization/79390)

2017-04-02 Thread Uros Bizjak
On Sat, Apr 1, 2017 at 2:20 PM, Jakub Jelinek wrote: > Hi! > > As discussed in the PR, in the following testcase we don't if-convert > with the generic (and many other) tuning, because we default to > --param max-rtl-if-conversion-insns=1 in most of the tunings. > The problem we

Re: [PATCH] On x86 allow if-conversion of more than one insn as long as there is at most one cmov (PR tree-optimization/79390)

2017-04-01 Thread Jakub Jelinek
On Sat, Apr 01, 2017 at 01:43:36PM -0500, Segher Boessenkool wrote: > Hi, > > On Sat, Apr 01, 2017 at 02:20:27PM +0200, Jakub Jelinek wrote: > > As discussed in the PR, in the following testcase we don't if-convert > > with the generic (and many other) tuning, because we default to > > --param

Re: [PATCH] On x86 allow if-conversion of more than one insn as long as there is at most one cmov (PR tree-optimization/79390)

2017-04-01 Thread Segher Boessenkool
Hi, On Sat, Apr 01, 2017 at 02:20:27PM +0200, Jakub Jelinek wrote: > As discussed in the PR, in the following testcase we don't if-convert > with the generic (and many other) tuning, because we default to > --param max-rtl-if-conversion-insns=1 in most of the tunings. > The problem we have is

[PATCH] On x86 allow if-conversion of more than one insn as long as there is at most one cmov (PR tree-optimization/79390)

2017-04-01 Thread Jakub Jelinek
Hi! As discussed in the PR, in the following testcase we don't if-convert with the generic (and many other) tuning, because we default to --param max-rtl-if-conversion-insns=1 in most of the tunings. The problem we have is with multiple cmov instructions, but in the testcase there is just one