[Bug middle-end/66916] New: [6 Regression] FAIL: gcc.target/arm/pr43920-2.c object-size text <= 54

[Bug middle-end/66916] New: [6 Regression] FAIL: gcc.target/arm/pr43920-2.c object-size text <= 54

[ktkachov at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66916

            Bug ID: 66916
           Summary: [6 Regression] FAIL: gcc.target/arm/pr43920-2.c
                    object-size text <= 54
           Product: gcc
           Version: 6.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: middle-end
          Assignee: unassigned at gcc dot gnu.org
          Reporter: ktkachov at gcc dot gnu.org
                CC: rguenth at gcc dot gnu.org
  Target Milestone: ---
             Build: arm*

Created attachment 36006
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=36006&action=edit
Failing assembly

After this commit I'm seeing the test fail on arm targets:

Author: rguenth <rguenth@138bc75d-0d04-0410-961f-82ee72b054a4>
Date:   Thu Jul 16 08:28:51 2015 +0000

    2015-07-16  Richard Biener  <rguenther@suse.de>

        * fold-const.c (fold_widened_comparison): Remove.
        (fold_sign_changed_comparison): Likewise.
        (fold_comparison): Move widened and sign-changed comparison
        simplification ...
        * match.pd: ... to patterns here.
        * generic-match-head.c: Include target.h.
        * gimple-match-head.c: Likewise.

        * gcc.dg/tree-ssa/pr21031.c: Adjust.


    git-svn-id: svn+ssh://gcc.gnu.org/svn/gcc/trunk@225861
138bc75d-0d04-0410-961f-82ee72b054a4


a sub+cmp sequence is not merged into a subs, causing the code size regression

[Bug middle-end/66916] [6 Regression] FAIL: gcc.target/arm/pr43920-2.c object-size text <= 54

[ktkachov at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66916

--- Comment #1 from ktkachov at gcc dot gnu.org ---
Created attachment 36007
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=36007&action=edit
good assembly before the offending commit

Attaching the good assembly before the commit

[Bug middle-end/66916] [6 Regression] FAIL: gcc.target/arm/pr43920-2.c object-size text <= 54

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66916

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |ASSIGNED
   Last reconfirmed|                            |2015-07-21
           Assignee|unassigned at gcc dot gnu.org      |rguenth at gcc dot gnu.org
   Target Milestone|---                         |6.0
     Ever confirmed|0                           |1

--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
Mine.  I'll have to think about a more general solution to handle condition
codes and insn fusing with branches and preceeding instructions.

[Bug middle-end/66916] [6 Regression] FAIL: gcc.target/arm/pr43920-2.c object-size text <= 54

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66916

--- Comment #3 from Richard Biener <rguenth at gcc dot gnu.org> ---
It's a combination of sign-changed compare and  X - Y CMP 0 to X CMP Y.  Bad
is

  <bb 4>:
  _9 = end_8 - start_6;
  length_10 = (size_t) _9;
  if (start_6 == end_8)

and I guess good was

  if (length_10 == 0)

where the sign-change actually didn't matter.  The single-use restriction
on the X - Y CMP 0 to X CMP Y doesn't help us because it has just a single
use - the conversion.  The sign-change and widen comparison optimization
has no such restriction though.  Adding that restriction avoids the above
transform and fixes the testcase.

[Bug middle-end/66916] [6 Regression] FAIL: gcc.target/arm/pr43920-2.c object-size text <= 54

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66916

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

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

--- Comment #4 from Richard Biener <rguenth at gcc dot gnu.org> ---
Should be fixed now.

[Bug middle-end/66916] [6 Regression] FAIL: gcc.target/arm/pr43920-2.c object-size text <= 54

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66916

--- Comment #5 from Richard Biener <rguenth at gcc dot gnu.org> ---
Author: rguenth
Date: Thu Jul 23 07:29:53 2015
New Revision: 226089

URL: https://gcc.gnu.org/viewcvs?rev=226089&root=gcc&view=rev
Log:
2015-07-23  Richard Biener  <rguenther@suse.de>

        PR middle-end/66916
        * match.pd: Guard widen and sign-change comparison simplification
        with single_use.

Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/match.pd