[Bug tree-optimization/107130] New: [13 Regression] wrong code with -Os -fno-tree-ccp -fno-tree-forwprop -fno-tree-fre

[Bug tree-optimization/107130] New: [13 Regression] wrong code with -Os -fno-tree-ccp -fno-tree-forwprop -fno-tree-fre

[zsojka at seznam dot cz]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107130

            Bug ID: 107130
           Summary: [13 Regression] wrong code with -Os -fno-tree-ccp
                    -fno-tree-forwprop -fno-tree-fre
           Product: gcc
           Version: 13.0
            Status: UNCONFIRMED
          Keywords: wrong-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

Created attachment 53653
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=53653&action=edit
reduced testcase

Output:
$ x86_64-pc-linux-gnu-gcc -Os -fno-tree-ccp -fno-tree-forwprop -fno-tree-fre
testcase.c
$ ./a.out 
Aborted

The generated code just aborts:
main:
        push    rax
        call    abort


$ x86_64-pc-linux-gnu-gcc -v
Using built-in specs.
COLLECT_GCC=/repo/gcc-trunk/binary-latest/bin/x86_64-pc-linux-gnu-gcc
COLLECT_LTO_WRAPPER=/repo/gcc-trunk/binary-trunk-r13-3023-20221003092645-g31d7c8bc263-checking-yes-rtl-df-extra-nobootstrap-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
--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-r13-3023-20221003092645-g31d7c8bc263-checking-yes-rtl-df-extra-nobootstrap-amd64
Thread model: posix
Supported LTO compression algorithms: zlib zstd
gcc version 13.0.0 20221003 (experimental) (GCC)

[Bug tree-optimization/107130] [13 Regression] wrong code with -Os -fno-tree-ccp -fno-tree-forwprop -fno-tree-fre since r13-2900-g001c60ccfeaf9a48

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107130

Martin Liška <marxin at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|                            |2022-10-03
             Status|UNCONFIRMED                 |NEW
     Ever confirmed|0                           |1
            Summary|[13 Regression] wrong code  |[13 Regression] wrong code
                   |with -Os -fno-tree-ccp      |with -Os -fno-tree-ccp
                   |-fno-tree-forwprop          |-fno-tree-forwprop
                   |-fno-tree-fre               |-fno-tree-fre since
                   |                            |r13-2900-g001c60ccfeaf9a48
                 CC|                            |aldyh at gcc dot gnu.org,
                   |                            |marxin at gcc dot gnu.org

--- Comment #1 from Martin Liška <marxin at gcc dot gnu.org> ---
Started with r13-2900-g001c60ccfeaf9a48.

[Bug tree-optimization/107130] [13 Regression] wrong code with -Os -fno-tree-ccp -fno-tree-forwprop -fno-tree-fre since r13-2900-g001c60ccfeaf9a48

[zsojka at seznam dot cz]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107130

--- Comment #2 from Zdenek Sojka <zsojka at seznam dot cz> ---
(In reply to Martin Liška from comment #1)
> Started with r13-2900-g001c60ccfeaf9a48.

Nice, so ffs() with a propagated constant argument now does a popcount()
instead?

[Bug tree-optimization/107130] [13 Regression] wrong code with -Os -fno-tree-ccp -fno-tree-forwprop -fno-tree-fre since r13-2900-g001c60ccfeaf9a48

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107130

--- Comment #3 from Martin Liška <marxin at gcc dot gnu.org> ---
> Nice, so ffs() with a propagated constant argument now does a popcount()
> instead?

It should not. The problem is op_cfn_popcount is shared by both popcount and
ffs, but the newly added transformation is valid only for popcount.

[Bug tree-optimization/107130] [13 Regression] wrong code with -Os -fno-tree-ccp -fno-tree-forwprop -fno-tree-fre since r13-2900-g001c60ccfeaf9a48

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107130

Martin Liška <marxin at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |13.0
           Priority|P3                          |P1

[Bug tree-optimization/107130] [13 Regression] wrong code with -Os -fno-tree-ccp -fno-tree-forwprop -fno-tree-fre since r13-2900-g001c60ccfeaf9a48

[aldyh at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107130

Aldy Hernandez <aldyh at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Assignee|unassigned at gcc dot gnu.org      |aldyh at gcc dot gnu.org

--- Comment #4 from Aldy Hernandez <aldyh at gcc dot gnu.org> ---
(In reply to Martin Liška from comment #3)
> > Nice, so ffs() with a propagated constant argument now does a popcount()
> > instead?
> 
> It should not. The problem is op_cfn_popcount is shared by both popcount and
> ffs, but the newly added transformation is valid only for popcount.

Ughh, indeed.  Thanks for the analysis Martin.

[Bug tree-optimization/107130] [13 Regression] wrong code with -Os -fno-tree-ccp -fno-tree-forwprop -fno-tree-fre since r13-2900-g001c60ccfeaf9a48

[aldyh at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107130

--- Comment #5 from Aldy Hernandez <aldyh at gcc dot gnu.org> ---
Created attachment 53656
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=53656&action=edit
untested patch

[Bug tree-optimization/107130] [13 Regression] wrong code with -Os -fno-tree-ccp -fno-tree-forwprop -fno-tree-fre since r13-2900-g001c60ccfeaf9a48

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107130

--- Comment #6 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Aldy Hernandez <aldyh@gcc.gnu.org>:

https://gcc.gnu.org/g:f50d103183c551c7f9f9f20efaf2ebbf83d5e99f

commit r13-3052-gf50d103183c551c7f9f9f20efaf2ebbf83d5e99f
Author: Aldy Hernandez <aldyh@redhat.com>
Date:   Mon Oct 3 17:31:32 2022 +0200

    [PR107130] range-ops: Separate out ffs and popcount optimizations.

    __builtin_popcount and __builtin_ffs were sharing the same range-ops
    entry, but the nonzero mask optimization is not valid for ffs.
    Separate them out into two entries.

            PR tree-optimization/107130

    gcc/ChangeLog:

            * gimple-range-op.cc (class cfn_popcount): Call op_cfn_ffs.
            (class cfn_ffs): New.
            (gimple_range_op_handler::maybe_builtin_call): Separate out
            CASE_CFN_FFS into its own case.

    gcc/testsuite/ChangeLog:

            * gcc.dg/tree-ssa/pr107130.c: New test.

[Bug tree-optimization/107130] [13 Regression] wrong code with -Os -fno-tree-ccp -fno-tree-forwprop -fno-tree-fre since r13-2900-g001c60ccfeaf9a48

[aldyh at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107130

Aldy Hernandez <aldyh at gcc dot gnu.org> changed:

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

--- Comment #7 from Aldy Hernandez <aldyh at gcc dot gnu.org> ---
fixed