[Bug middle-end/116601] New: [15 Regression] during GIMPLE pass: fab ICE: verify_gimple failed: statement marked for throw, but doesn't with -O -fnon-call-exceptions

[Bug middle-end/116601] New: [15 Regression] during GIMPLE pass: fab ICE: verify_gimple failed: statement marked for throw, but doesn't with -O -fnon-call-exceptions

[zsojka at seznam dot cz]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116601

            Bug ID: 116601
           Summary: [15 Regression] during GIMPLE pass: fab ICE:
                    verify_gimple failed: statement marked for throw, but
                    doesn't with -O -fnon-call-exceptions
           Product: gcc
           Version: 15.0
            Status: UNCONFIRMED
          Keywords: ice-on-valid-code
          Severity: normal
          Priority: P3
         Component: middle-end
          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 59050
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=59050&action=edit
auto-reduced testcase

Compiler output:
$ x86_64-pc-linux-gnu-gcc -O -fnon-call-exceptions testcase.C 
testcase.C: In function 'void ShowVehicleRefitWindow()':
testcase.C:22:6: error: statement marked for throw, but doesn't
   22 | void ShowVehicleRefitWindow() { new RefitWindow; }
      |      ^~~~~~~~~~~~~~~~~~~~~~
# .MEM_11 = VDEF <.MEM_10>
__trans_tmp_2 = {};
during GIMPLE pass: fab
testcase.C:22:6: internal compiler error: verify_gimple failed
0x2eeb08e internal_error(char const*, ...)
        /repo/gcc-trunk/gcc/diagnostic-global-context.cc:492
0x18fcded verify_gimple_in_cfg(function*, bool, bool)
        /repo/gcc-trunk/gcc/tree-cfg.cc:5681
0x176da14 execute_function_todo
        /repo/gcc-trunk/gcc/passes.cc:2089
0x176df6e execute_todo
        /repo/gcc-trunk/gcc/passes.cc:2143
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-r15-3443-20240904104428-gda3a2985fff-checking-yes-rtl-df-extra-nobootstrap-amd64/bin/../libexec/gcc/x86_64-pc-linux-gnu/15.0.0/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
--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 --enable-libsanitizer
--disable-libstdcxx-pch
--prefix=/repo/gcc-trunk//binary-trunk-r15-3443-20240904104428-gda3a2985fff-checking-yes-rtl-df-extra-nobootstrap-amd64
Thread model: posix
Supported LTO compression algorithms: zlib zstd
gcc version 15.0.0 20240903 (experimental) (GCC)

[Bug tree-optimization/116601] [15 Regression] during GIMPLE pass: fab ICE: verify_gimple failed: statement marked for throw, but doesn't with -O -fnon-call-exceptions

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116601

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |ASSIGNED
          Component|middle-end                  |tree-optimization
     Ever confirmed|0                           |1
   Last reconfirmed|                            |2024-09-04
   Target Milestone|---                         |15.0
           Assignee|unassigned at gcc dot gnu.org      |pinskia at gcc dot gnu.org

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Confirmed.


  <bb 2> [local count: 1073741824]:
  *__val_6(D) = {};

  <bb 3> [local count: 1073741824]:
  __trans_tmp_2 = MEM[(const struct RefitOption &)__val_6(D)];
  goto <bb 5>; [100.00%]


Into:
  <bb 2> [local count: 1073741824]:
  *__val_6(D) = {};

  <bb 3> [local count: 1073741824]:
  __trans_tmp_2 = {};
  goto <bb 5>; [100.00%]


The difference between GCC 14 and 15 comes from gimplifier.

Anyways I am going to fix this.

[Bug tree-optimization/116601] [15 Regression] during GIMPLE pass: fab ICE: verify_gimple failed: statement marked for throw, but doesn't with -O -fnon-call-exceptions

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116601

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
The bug is in optimize_memcpy or maybe after the call to optimize_memcpy. It
has been a latent issue since r7-5443-g7b45d0dfeb5f85 .

[Bug tree-optimization/116601] [15 Regression] during GIMPLE pass: fab ICE: verify_gimple failed: statement marked for throw, but doesn't with -O -fnon-call-exceptions

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116601

--- Comment #3 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
The fix which I am testing:
```
diff --git a/gcc/tree-ssa-ccp.cc b/gcc/tree-ssa-ccp.cc
index 44711018e0e..3cd385f476b 100644
--- a/gcc/tree-ssa-ccp.cc
+++ b/gcc/tree-ssa-ccp.cc
@@ -4325,8 +4325,15 @@ pass_fold_builtins::execute (function *fun)
           if (gimple_code (stmt) != GIMPLE_CALL)
            {
              if (gimple_assign_load_p (stmt) && gimple_store_p (stmt))
-               optimize_memcpy (&i, gimple_assign_lhs (stmt),
-                                gimple_assign_rhs1 (stmt), NULL_TREE);
+               {
+                 optimize_memcpy (&i, gimple_assign_lhs (stmt),
+                                  gimple_assign_rhs1 (stmt), NULL_TREE);
+                 old_stmt = stmt;
+                 stmt = gsi_stmt (i);
+                 if (maybe_clean_or_replace_eh_stmt (old_stmt, stmt)
+                     && gimple_purge_dead_eh_edges (bb))
+                   cfg_changed = true;
+               }
              gsi_next (&i);
              continue;
            }

```

[Bug tree-optimization/116601] [15 Regression] during GIMPLE pass: fab ICE: verify_gimple failed: statement marked for throw, but doesn't with -O -fnon-call-exceptions

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116601

--- Comment #4 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Created attachment 59051
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=59051&action=edit
Cleaned up testcase

`-O1 -fnon-call-exceptions`

This cleaned up testcase fails since GCC 7 even.

[Bug tree-optimization/116601] [12/13/14/15 Regression] during GIMPLE pass: fab ICE: verify_gimple failed: statement marked for throw, but doesn't with -O -fnon-call-exceptions

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116601

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[15 Regression] during      |[12/13/14/15 Regression]
                   |GIMPLE pass: fab ICE:       |during GIMPLE pass: fab
                   |verify_gimple failed:       |ICE: verify_gimple failed:
                   |statement marked for throw, |statement marked for throw,
                   |but doesn't with -O         |but doesn't with -O
                   |-fnon-call-exceptions       |-fnon-call-exceptions
      Known to fail|                            |7.1.0
   Target Milestone|15.0                        |12.5
      Known to work|14.2.1                      |6.1.0

[Bug tree-optimization/116601] [12/13/14/15 Regression] during GIMPLE pass: fab ICE: verify_gimple failed: statement marked for throw, but doesn't with -O -fnon-call-exceptions

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116601

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |patch
                URL|                            |https://gcc.gnu.org/piperma
                   |                            |il/gcc-patches/2024-Septemb
                   |                            |er/662327.html

--- Comment #5 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Patch posted:
https://gcc.gnu.org/pipermail/gcc-patches/2024-September/662327.html

[Bug tree-optimization/116601] [12/13/14/15 Regression] during GIMPLE pass: fab ICE: verify_gimple failed: statement marked for throw, but doesn't with -O -fnon-call-exceptions

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116601

--- Comment #6 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Created attachment 59065
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=59065&action=edit
Patch which moves this folding to a different location

So in the end, it is better to move this optimization to fold_stmt. For
branches my original patch seems like the best.

[Bug tree-optimization/116601] [12/13/14/15 Regression] during GIMPLE pass: fab ICE: verify_gimple failed: statement marked for throw, but doesn't with -O -fnon-call-exceptions

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116601

--- Comment #7 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The trunk branch has been updated by Andrew Pinski <pinskia@gcc.gnu.org>:

https://gcc.gnu.org/g:2067df800df930dd13aa20a826373bb27cfdd420

commit r15-3544-g2067df800df930dd13aa20a826373bb27cfdd420
Author: Andrew Pinski <quic_apinski@quicinc.com>
Date:   Fri Sep 6 12:29:26 2024 -0700

    gimple-fold: Move optimizing memcpy to memset to fold_stmt from fab

    I noticed this folding inside fab could be done else where and could
    even improve inlining decisions and a few other things so let's
    move it to fold_stmt.
    It also fixes PR 116601 because places which call fold_stmt already
    have to deal with the stmt becoming a non-throw statement.

    For the fix for PR 116601 on the branches should be the original patch
    rather than a backport of this one.

    Bootstrapped and tested on x86_64-linux-gnu.

            PR tree-optimization/116601

    gcc/ChangeLog:

            * gimple-fold.cc (optimize_memcpy_to_memset): Move
            from tree-ssa-ccp.cc and rename. Also return true
            if the optimization happened.
            (gimple_fold_builtin_memory_op): Call
            optimize_memcpy_to_memset.
            (fold_stmt_1): Call optimize_memcpy_to_memset for
            load/store copies.
            * tree-ssa-ccp.cc (optimize_memcpy): Delete.
            (pass_fold_builtins::execute): Remove code that
            calls optimize_memcpy.

    gcc/testsuite/ChangeLog:

            * gcc.dg/pr78408-1.c: Adjust dump scan to match where
            the optimization now happens.
            * g++.dg/torture/except-2.C: New test.

    Signed-off-by: Andrew Pinski <quic_apinski@quicinc.com>

[Bug tree-optimization/116601] [12/13/14 Regression] during GIMPLE pass: fab ICE: verify_gimple failed: statement marked for throw, but doesn't with -O -fnon-call-exceptions

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116601

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
      Known to fail|15.0                        |
            Summary|[12/13/14/15 Regression]    |[12/13/14 Regression]
                   |during GIMPLE pass: fab     |during GIMPLE pass: fab
                   |ICE: verify_gimple failed:  |ICE: verify_gimple failed:
                   |statement marked for throw, |statement marked for throw,
                   |but doesn't with -O         |but doesn't with -O
                   |-fnon-call-exceptions       |-fnon-call-exceptions
      Known to work|                            |15.0

--- Comment #8 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Fixed on the trunk so far by moving the optimization to a better location.