[Bug c++/96163] New: another ice in pass slp

[Bug c++/96163] New: another ice in pass slp

[dcb314 at hotmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96163

            Bug ID: 96163
           Summary: another ice in pass slp
           Product: gcc
           Version: unknown
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: dcb314 at hotmail dot com
  Target Milestone: ---

Created attachment 48862
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=48862&action=edit
C++ source code

For the attached C++ source code with flag -O3, recent gcc trunk does this:

/home/dcb/gcc/results.20200630/bin/g++
/home/dcb/gcc/results.20200701/bin/g++
bug630.cc: In function ‘void cm()’:
bug630.cc:139:6: error: definition in block 2 follows the use
  139 | void cm() {
      |      ^~
for SSA_NAME: vect_cst__16 in statement:
vect__2.37_17 = _14 / vect_cst__16;
during GIMPLE pass: slp
bug630.cc:139:6: internal compiler error: verify_ssa failed
0x13fb41b verify_ssa(bool, bool)
        ../../trunk.git/gcc/tree-ssa.c:1208
0x108dbf5 execute_function_todo
        ../../trunk.git/gcc/passes.c:1992

Trunk dated 20200630 is fine, 20200701 not.

[Bug tree-optimization/96163] another ice in pass slp

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96163

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |ASSIGNED
            Version|unknown                     |11.0
     Ever confirmed|0                           |1
          Component|c++                         |tree-optimization
           Assignee|unassigned at gcc dot gnu.org      |rguenth at gcc dot gnu.org
   Last reconfirmed|                            |2020-07-13

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

[Bug tree-optimization/96163] another ice in pass slp

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96163

--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
OK, so the issue is that the representation for pre-existing vectors leaves
us with vector defs outside of the region (and not immediately before the
region begin) which then confuses the logic in vect_stmt_dominates_stmt_p.

I need to think a bit about this.

[Bug tree-optimization/96163] another ice in pass slp

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96163

--- Comment #3 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Richard Biener <rguenth@gcc.gnu.org>:

https://gcc.gnu.org/g:c4facd483de2d5b6b4d20e6ad74c85b9f1eba41a

commit r11-2062-gc4facd483de2d5b6b4d20e6ad74c85b9f1eba41a
Author: Richard Biener <rguenther@suse.de>
Date:   Mon Jul 13 12:41:35 2020 +0200

    tree-optimization/96163 - fix placement issue with SLP and vectors

    This avoids placing stmts beyond the vectorizer region begin which
    confuses vect_stmt_dominates_stmt_p.

    2020-07-13  Richard Biener  <rguenther@suse.de>

            PR tree-optimization/96163
            * tree-vect-slp.c (vect_schedule_slp_instance): Put new stmts
            at least after region begin.

            * g++.dg/vect/pr96163.cc: New testcase.

[Bug tree-optimization/96163] another ice in pass slp

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96163

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

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

--- Comment #4 from Richard Biener <rguenth at gcc dot gnu.org> ---
Fixed.