[Bug middle-end/63790] New: [5 Regression] Tests XFAILed because of the match-and-simplify merge

[Bug middle-end/63790] New: [5 Regression] Tests XFAILed because of the match-and-simplify merge

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63790

            Bug ID: 63790
           Summary: [5 Regression] Tests XFAILed because of the
                    match-and-simplify merge
           Product: gcc
           Version: 5.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: middle-end
          Assignee: unassigned at gcc dot gnu.org
          Reporter: rguenth at gcc dot gnu.org

g++.dg/cpp0x/constexpr-reinterpret1.C
gcc.dg/tree-ssa/pr21031.c

[Bug middle-end/63790] [5 Regression] Tests XFAILed because of the match-and-simplify merge

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63790

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |xfail
             Status|UNCONFIRMED                 |ASSIGNED
   Last reconfirmed|                            |2014-11-09
           Assignee|unassigned at gcc dot gnu.org      |rguenth at gcc dot gnu.org
   Target Milestone|---                         |5.0
     Ever confirmed|0                           |1

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
Mine.

[Bug middle-end/63790] [5 Regression] Tests XFAILed because of the match-and-simplify merge

[ppalka at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63790

--- Comment #2 from ppalka at gcc dot gnu.org ---
Author: ppalka
Date: Mon Nov 17 02:01:36 2014
New Revision: 217638

URL: https://gcc.gnu.org/viewcvs?rev=217638&root=gcc&view=rev
Log:
Always combine comparisons or conversions from booleans.

2014-11-16  Patrick Palka  <ppalka@gcc.gnu.org>

gcc/
    PR middle-end/63790
    * tree-ssa-forwprop.c (forward_propagate_into_comparison_1):
    Always combine comparisons or conversions from booleans.

gcc/testsuite/
    PR middle-end/63790
    * gcc.dg/tree-ssa/pr21031.c: Drop XFAIL.
    * gcc.dg/tree-ssa/forwprop-29.c: New test.


Added:
    trunk/gcc/testsuite/gcc.dg/tree-ssa/forwprop-29.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/testsuite/gcc.dg/tree-ssa/pr21031.c
    trunk/gcc/tree-ssa-forwprop.c

[Bug middle-end/63790] [5 Regression] Tests XFAILed because of the match-and-simplify merge

[fxcoudert at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63790

Francois-Xavier Coudert <fxcoudert at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |fxcoudert at gcc dot gnu.org

--- Comment #3 from Francois-Xavier Coudert <fxcoudert at gcc dot gnu.org> ---
forwprop-29.c fails on x86_64-apple-darwin14, because it does not emit an error
for line 21.

[Bug middle-end/63790] [5 Regression] Tests XFAILed because of the match-and-simplify merge

[rguenther at suse dot de]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63790

--- Comment #4 from rguenther at suse dot de <rguenther at suse dot de> ---
On Tue, 18 Nov 2014, fxcoudert at gcc dot gnu.org wrote:

> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63790
> 
> Francois-Xavier Coudert <fxcoudert at gcc dot gnu.org> changed:
> 
>            What    |Removed                     |Added
> ----------------------------------------------------------------------------
>                  CC|                            |fxcoudert at gcc dot gnu.org
> 
> --- Comment #3 from Francois-Xavier Coudert <fxcoudert at gcc dot gnu.org> ---
> forwprop-29.c fails on x86_64-apple-darwin14, because it does not emit an error
> for line 21.

It expects both functions to be inlined.  Can't reproduce with darwin
specials (-fPIC -fno-ipa-icf).  Can you upload 088t.dom1 dump?

[Bug middle-end/63790] [5 Regression] Tests XFAILed because of the match-and-simplify merge

[fxcoudert at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63790

--- Comment #5 from Francois-Xavier Coudert <fxcoudert at gcc dot gnu.org> ---
With -O2 -fno-ipa-icf, the expected error shows up. With only -O2, it does not.
I attach the 088t.dom1 dump for both cases.

[Bug middle-end/63790] [5 Regression] Tests XFAILed because of the match-and-simplify merge

[fxcoudert at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63790

--- Comment #6 from Francois-Xavier Coudert <fxcoudert at gcc dot gnu.org> ---
Created attachment 34021
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=34021&action=edit
dom1 dump with -O2

[Bug middle-end/63790] [5 Regression] Tests XFAILed because of the match-and-simplify merge

[fxcoudert at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63790

--- Comment #7 from Francois-Xavier Coudert <fxcoudert at gcc dot gnu.org> ---
Created attachment 34022
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=34022&action=edit
dom1 dump with -O2 -fno-ipa-icf

[Bug middle-end/63790] [5 Regression] Tests XFAILed because of the match-and-simplify merge

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63790

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |mliska at suse dot cz

--- Comment #8 from Richard Biener <rguenth at gcc dot gnu.org> ---
(In reply to Francois-Xavier Coudert from comment #5)
> With -O2 -fno-ipa-icf, the expected error shows up. With only -O2, it does
> not.
> I attach the 088t.dom1 dump for both cases.

Interesting that IPA-ICF doesn't catch the opportunity on x86_64 ... Martin?
Ah, it seems we unify it but still inline it twice while on dawrin we
unify it but do not inline it (but CSE one call later).

But yes, adding -fno-ipa-icf to dg-options is fine with me (and pre-approved).

[Bug middle-end/63790] [5 Regression] Tests XFAILed because of the match-and-simplify merge

[dominiq at lps dot ens.fr]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63790

--- Comment #9 from Dominique d'Humieres <dominiq at lps dot ens.fr> ---
> Interesting that IPA-ICF doesn't catch the opportunity on x86_64 ... Martin?

IPA-ICF is badly broken on darwin.

[Bug middle-end/63790] [5 Regression] Tests XFAILed because of the match-and-simplify merge

[fxcoudert at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63790

--- Comment #10 from Francois-Xavier Coudert <fxcoudert at gcc dot gnu.org> ---
(In reply to Richard Biener from comment #8)
> But yes, adding -fno-ipa-icf to dg-options is fine with me (and
> pre-approved).

I'm a bit worried that by manually disabling ipa-icf we paper over the bugs on
darwin, rather than get them fixed in the first place.

[Bug middle-end/63790] [5 Regression] Tests XFAILed because of the match-and-simplify merge

[danglin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63790

John David Anglin <danglin at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |danglin at gcc dot gnu.org

--- Comment #11 from John David Anglin <danglin at gcc dot gnu.org> ---
Also fails on hppa2.0w-hp-hpux11.11.

[Bug middle-end/63790] [5 Regression] Tests XFAILed because of the match-and-simplify merge

[mpolacek at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63790
Bug 63790 depends on bug 63862, which changed state.

Bug 63862 Summary: C frontend converts shift-count to int while standard wants integer promotions
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63862

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|---                         |FIXED

[Bug middle-end/63790] [5 Regression] Tests XFAILed because of the match-and-simplify merge

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63790

--- Comment #12 from Richard Biener <rguenth at gcc dot gnu.org> ---
As far as I know we're clean but on the XFAIL of
g++.dg/cpp0x/constexpr-reinterpret1.C which was acked by Jason though.

[Bug middle-end/63790] [5 Regression] Tests XFAILed because of the match-and-simplify merge

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63790

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|---                         |FIXED

--- Comment #13 from Richard Biener <rguenth at gcc dot gnu.org> ---
Thus fixed.