[Bug tree-optimization/107686] New: [12/13 Regression] ICE: verify_gimple failed (invalid types in nop conversion) with _Decimal64 vector

[Bug tree-optimization/107686] New: [12/13 Regression] ICE: verify_gimple failed (invalid types in nop conversion) with _Decimal64 vector

[zsojka at seznam dot cz]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107686

            Bug ID: 107686
           Summary: [12/13 Regression] ICE: verify_gimple failed (invalid
                    types in nop conversion) with _Decimal64 vector
           Product: gcc
           Version: 13.0
            Status: UNCONFIRMED
          Keywords: ice-on-valid-code
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: zsojka at seznam dot cz
  Target Milestone: ---
              Host: x86_64-pc-linux-gnu
            Target: x86_64-pc-linux-gnu

Compiler output:
$ x86_64-pc-linux-gnu-gcc -O -mavx2 testcase.c 
testcase.c: In function 'foo':
testcase.c:9:1: error: invalid types in nop conversion
    9 | foo (void)
      | ^~~
_Decimal64
int
_33 = (_Decimal64) _1;
testcase.c:9:1: error: invalid types in nop conversion
_Decimal64
int
_26 = (_Decimal64) _3;
testcase.c:9:1: error: invalid types in nop conversion
_Decimal64
int
_19 = (_Decimal64) _15;
testcase.c:9:1: error: invalid types in nop conversion
_Decimal64
int
_12 = (_Decimal64) _22;
testcase.c:9:1: error: invalid types in nop conversion
_Decimal64
int
_61 = (_Decimal64) _29;
testcase.c:9:1: error: invalid types in nop conversion
_Decimal64
int
_54 = (_Decimal64) _36;
testcase.c:9:1: error: invalid types in nop conversion
_Decimal64
int
_47 = (_Decimal64) _43;
testcase.c:9:1: error: invalid types in nop conversion
_Decimal64
int
_40 = (_Decimal64) _50;
during GIMPLE pass: forwprop
testcase.c:9:1: internal compiler error: verify_gimple failed
0x142be1b verify_gimple_in_cfg(function*, bool, bool)
        /repo/gcc-trunk/gcc/tree-cfg.cc:5658
0x12d4590 execute_function_todo
        /repo/gcc-trunk/gcc/passes.cc:2091
0x12d4ace execute_todo
        /repo/gcc-trunk/gcc/passes.cc:2145
Please submit a full bug report, with preprocessed source (by using
-freport-bug).
Please include the complete backtrace with any bug report.
See <https://gcc.gnu.org/bugs/> for instructions.

$ x86_64-pc-linux-gnu-gcc -v
Using built-in specs.
COLLECT_GCC=/repo/gcc-trunk/binary-latest-amd64/bin/x86_64-pc-linux-gnu-gcc
COLLECT_LTO_WRAPPER=/repo/gcc-trunk/binary-trunk-r13-4023-20221114155950-g2b85d759dae-checking-yes-rtl-df-extra-amd64/bin/../libexec/gcc/x86_64-pc-linux-gnu/13.0.0/lto-wrapper
Target: x86_64-pc-linux-gnu
Configured with: /repo/gcc-trunk//configure --enable-languages=c,c++
--enable-valgrind-annotations --disable-nls --enable-checking=yes,rtl,df,extra
--with-cloog --with-ppl --with-isl --build=x86_64-pc-linux-gnu
--host=x86_64-pc-linux-gnu --target=x86_64-pc-linux-gnu
--with-ld=/usr/bin/x86_64-pc-linux-gnu-ld
--with-as=/usr/bin/x86_64-pc-linux-gnu-as --disable-libstdcxx-pch
--prefix=/repo/gcc-trunk//binary-trunk-r13-4023-20221114155950-g2b85d759dae-checking-yes-rtl-df-extra-amd64
Thread model: posix
Supported LTO compression algorithms: zlib zstd
gcc version 13.0.0 20221114 (experimental) (GCC)

[Bug tree-optimization/107686] [12/13 Regression] ICE: verify_gimple failed (invalid types in nop conversion) with _Decimal64 vector

[zsojka at seznam dot cz]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107686

--- Comment #1 from Zdenek Sojka <zsojka at seznam dot cz> ---
Created attachment 53896
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=53896&action=edit
reduced testcase

[Bug tree-optimization/107686] [12/13 Regression] ICE: verify_gimple failed (invalid types in nop conversion) with _Decimal64 vector

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107686

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---

  vector(4) long intD.8 _10;

  _10 = [vec_unpack_lo_expr] u.0_2;
  _11 = [vec_unpack_hi_expr] u.0_2;
  # VUSE <.MEM_8(D)>
  d.1_4 = dD.3174;
  _12 = BIT_FIELD_REF <_10, 64, 0>;


  _1 = BIT_FIELD_REF <uD.3173, 32, 96>;
  _33 = (_Decimal64D.41) _1;

[Bug tree-optimization/107686] [12/13 Regression] ICE: verify_gimple failed (invalid types in nop conversion) with _Decimal64 vector

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107686

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2022-11-14
     Ever confirmed|0                           |1
   Target Milestone|---                         |12.3

--- Comment #3 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Confirmed.

This code I think is broken:

2724d1bba6b3 gcc/tree-ssa-forwprop.c  (Richard Biener        2021-07-30
11:06:50 +0200 3205)      /* Perform scalar promotion.  */
2724d1bba6b3 gcc/tree-ssa-forwprop.c  (Richard Biener        2021-07-30
11:06:50 +0200 3206)      new_stmt = gimple_build_assign (gimple_assign_lhs
(use_stmt),
2724d1bba6b3 gcc/tree-ssa-forwprop.c  (Richard Biener        2021-07-30
11:06:50 +0200 3207)                                      NOP_EXPR, tem);


So this is most likely introduced by r12-2728-g2724d1bba6b364 .

[Bug tree-optimization/107686] [12/13 Regression] ICE: verify_gimple failed (invalid types in nop conversion) with _Decimal64 vector

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107686

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
           Assignee|unassigned at gcc dot gnu.org      |rguenth at gcc dot gnu.org

--- Comment #4 from Richard Biener <rguenth at gcc dot gnu.org> ---
I have a patch.

[Bug tree-optimization/107686] [12/13 Regression] ICE: verify_gimple failed (invalid types in nop conversion) with _Decimal64 vector

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107686

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P2

[Bug tree-optimization/107686] [12/13 Regression] ICE: verify_gimple failed (invalid types in nop conversion) with _Decimal64 vector

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107686

--- Comment #5 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:246bbdaa5f536b7a199dda9860c473137f40d622

commit r13-4102-g246bbdaa5f536b7a199dda9860c473137f40d622
Author: Richard Biener <rguenther@suse.de>
Date:   Wed Nov 16 15:27:13 2022 +0100

    tree-optimization/107686 - fix bitfield ref through vec_unpack optimization

    The following propely restricts the bitfield access to integral types
    when we look through VEC_UNPACK with the intent to emit a widening
    conversion.

            PR tree-optimization/107686
            * tree-ssa-forwprop.cc (optimize_vector_load): Restrict
            VEC_UNPACK support to integral typed bitfield refs.

            * gcc.dg/pr107686.c: New testcase.

[Bug tree-optimization/107686] [12 Regression] ICE: verify_gimple failed (invalid types in nop conversion) with _Decimal64 vector

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107686

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
      Known to fail|13.0                        |
            Summary|[12/13 Regression] ICE:     |[12 Regression] ICE:
                   |verify_gimple failed        |verify_gimple failed
                   |(invalid types in nop       |(invalid types in nop
                   |conversion) with _Decimal64 |conversion) with _Decimal64
                   |vector                      |vector
      Known to work|                            |13.0

--- Comment #6 from Richard Biener <rguenth at gcc dot gnu.org> ---
Fixed on trunk sofar.

[Bug tree-optimization/107686] [12 Regression] ICE: verify_gimple failed (invalid types in nop conversion) with _Decimal64 vector

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107686

--- Comment #7 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-12 branch has been updated by Richard Biener
<rguenth@gcc.gnu.org>:

https://gcc.gnu.org/g:232305bd0c8dd37b17b815864e730aa630714606

commit r12-8976-g232305bd0c8dd37b17b815864e730aa630714606
Author: Richard Biener <rguenther@suse.de>
Date:   Wed Nov 16 15:27:13 2022 +0100

    tree-optimization/107686 - fix bitfield ref through vec_unpack optimization

    The following propely restricts the bitfield access to integral types
    when we look through VEC_UNPACK with the intent to emit a widening
    conversion.

            PR tree-optimization/107686
            * tree-ssa-forwprop.cc (optimize_vector_load): Restrict
            VEC_UNPACK support to integral typed bitfield refs.

            * gcc.dg/pr107686.c: New testcase.

    (cherry picked from commit 246bbdaa5f536b7a199dda9860c473137f40d622)

[Bug tree-optimization/107686] [12 Regression] ICE: verify_gimple failed (invalid types in nop conversion) with _Decimal64 vector

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107686

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
      Known to fail|12.2.1                      |12.2.0
      Known to work|                            |12.2.1
         Resolution|---                         |FIXED
             Status|ASSIGNED                    |RESOLVED

--- Comment #8 from Richard Biener <rguenth at gcc dot gnu.org> ---
Fixed.