[Bug tree-optimization/113371] New: [14 Regression] ICE: verify_ssa failed: PHI node with wrong VUSE on edge from BB 19 with -O -march=silvermont -ftree-vectorize

[Bug tree-optimization/113371] New: [14 Regression] ICE: verify_ssa failed: PHI node with wrong VUSE on edge from BB 19 with -O -march=silvermont -ftree-vectorize

[zsojka at seznam dot cz]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113371

            Bug ID: 113371
           Summary: [14 Regression] ICE: verify_ssa failed: PHI node with
                    wrong VUSE on edge from BB 19 with -O
                    -march=silvermont -ftree-vectorize
           Product: gcc
           Version: 14.0
            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

Created attachment 57067
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=57067&action=edit
reduced testcase

Compiler output:
$ x86_64-pc-linux-gnu-gcc -O -march=silvermont -ftree-vectorize testcase.c 
testcase.c: In function 'BN_uadd':
testcase.c:4:1: error: PHI node with wrong VUSE on edge from BB 19
    4 | BN_uadd (int dif, long t1)
      | ^~~~~~~
.MEM_125 = PHI <.MEM_8(D)(19)>
expected .MEM_45
during GIMPLE pass: vect
testcase.c:4:1: internal compiler error: verify_ssa failed
0x177532f verify_ssa(bool, bool)
        /repo/gcc-trunk/gcc/tree-ssa.cc:1203
0x13c8545 execute_function_todo
        /repo/gcc-trunk/gcc/passes.cc:2095
0x13c89ae execute_todo
        /repo/gcc-trunk/gcc/passes.cc:2142
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-r14-7215-20240112190107-g8b447fa89d5-checking-yes-rtl-df-extra-nobootstrap-amd64/bin/../libexec/gcc/x86_64-pc-linux-gnu/14.0.1/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
--disable-bootstrap --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-r14-7215-20240112190107-g8b447fa89d5-checking-yes-rtl-df-extra-nobootstrap-amd64
Thread model: posix
Supported LTO compression algorithms: zlib zstd
gcc version 14.0.1 20240112 (experimental) (GCC)

[Bug tree-optimization/113371] [14 Regression] ICE: verify_ssa failed: PHI node with wrong VUSE on edge from BB 19 with -O -march=silvermont -ftree-vectorize

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113371

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |14.0
                 CC|                            |pinskia at gcc dot gnu.org

[Bug tree-optimization/113371] [14 Regression] ICE: verify_ssa failed: PHI node with wrong VUSE on edge from BB 19 with -O -march=silvermont -ftree-vectorize

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113371

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2024-01-13
     Ever confirmed|0                           |1

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Confirmed but note the reduced testcase depends on an uninitialized variable.

[Bug tree-optimization/113371] [14 Regression] ICE: verify_ssa failed: PHI node with wrong VUSE on edge from BB 19 with -O -march=silvermont -ftree-vectorize

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113371

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P1

--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
Confirmed, it's a recent regression.

[Bug tree-optimization/113371] [14 Regression] ICE: verify_ssa failed: PHI node with wrong VUSE on edge from BB 19 with -O -march=silvermont -ftree-vectorize

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113371

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

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

--- Comment #3 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Like in PR113374, started to ICE with one message in
r14-7194-g6cb155a6cf314232248a12bdd395ed4151ae5a28, another one in
r14-7195-g411de96dbf2bdafc7a90ebbfc63e68afd6388d29 and yet another one
r14-7196-g99c0a540d6689ede068f9ba98af6f38c3cd71362 which is what we get until
latest trunk.

[Bug tree-optimization/113371] [14 Regression] ICE: verify_ssa failed: PHI node with wrong VUSE on edge from BB 19 with -O -march=silvermont -ftree-vectorize

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113371

--- Comment #4 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
The https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113364#c9 patch doesn't fix
this.

[Bug tree-optimization/113371] [14 Regression] ICE: verify_ssa failed: PHI node with wrong VUSE on edge from BB 19 with -O -march=silvermont -ftree-vectorize

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113371

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

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

--- Comment #5 from Richard Biener <rguenth at gcc dot gnu.org> ---
I think we're doing sth unsupported - prologue peeling for
LOOP_VINFO_EARLY_BREAKS_VECT_PEELED.  The problem is that we only late
decide whether we do vect_use_loop_mask_for_alignment_p and I'm not sure
whether that variant works.

Testing a simple patch.

[Bug tree-optimization/113371] [14 Regression] ICE: verify_ssa failed: PHI node with wrong VUSE on edge from BB 19 with -O -march=silvermont -ftree-vectorize

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113371

--- Comment #6 from GCC 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:3359942417b02de88ae84d50aac232ac01ff9e15

commit r14-8175-g3359942417b02de88ae84d50aac232ac01ff9e15
Author: Richard Biener <rguenther@suse.de>
Date:   Tue Jan 16 13:09:27 2024 +0100

    tree-optimization/113371 - avoid prologue peeling for peeled early exits

    The following avoids prologue peeling when doing early exit
    vectorization with the IV exit before the early exit.  That's because
    we it invalidates the invariant that the effective latch of the loop
    is empty causing wrong continuation to the main loop.  In particular
    this is prone to break virtual SSA form.

            PR tree-optimization/113371
            * tree-vect-data-refs.cc (vect_enhance_data_refs_alignment):
            Do not peel when LOOP_VINFO_EARLY_BREAKS_VECT_PEELED.
            * tree-vect-loop-manip.cc (vect_do_peeling): Assert we do
            not perform prologue peeling when
LOOP_VINFO_EARLY_BREAKS_VECT_PEELED.

            * gcc.dg/vect/pr113371.c: New testcase.

[Bug tree-optimization/113371] [14 Regression] ICE: verify_ssa failed: PHI node with wrong VUSE on edge from BB 19 with -O -march=silvermont -ftree-vectorize

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113371

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

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

--- Comment #7 from Richard Biener <rguenth at gcc dot gnu.org> ---
Fixed.