[Bug analyzer/106551] New: [13 Regression] dup2 causes -fanalyser ICE in valid_to_unchecked_state, at analyzer/sm-fd.cc:751

[Bug analyzer/106551] New: [13 Regression] dup2 causes -fanalyser ICE in valid_to_unchecked_state, at analyzer/sm-fd.cc:751

[slyfox at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=106551

            Bug ID: 106551
           Summary: [13 Regression] dup2 causes -fanalyser ICE in
                    valid_to_unchecked_state, at analyzer/sm-fd.cc:751
           Product: gcc
           Version: 13.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: analyzer
          Assignee: dmalcolm at gcc dot gnu.org
          Reporter: slyfox at gcc dot gnu.org
  Target Milestone: ---

This week's gcc started ICEing on libpipeline-1.5.4 upstream package.

Minimal reproducer turned up by cvise:

    $ cat a.c.c
    extern int m;
    int dup2(int, int);
    void error(void);
    void max_active_pipelines_pdes(void) {
      if (dup2(m, 1))
        error();
    }

$ gcc -g -O2 -Wall  -c a.c.c -fanalyzer
during IPA pass: analyzer
a.c.c: In function 'max_active_pipelines_pdes':
a.c.c:5:7: internal compiler error: in valid_to_unchecked_state, at
analyzer/sm-fd.cc:751
    5 |   if (dup2(m, 1))
      |       ^~~~~~~~~~
0x1ef7d74 diagnostic_impl(rich_location*, diagnostic_metadata const*, int, char
const*, __va_list_tag (*) [1], diagnostic_t)
        ???:0
0x1ef89c6 internal_error(char const*, ...)
        ???:0
0x9023d4 fancy_abort(char const*, int, char const*)
        ???:0
0x85baa7 ana::(anonymous
namespace)::fd_state_machine::check_for_dup(ana::sm_context*, ana::supernode
const*, gimple const*, gcall const*, tree_node*, ana::(anonymous
namespace)::dup) const [clone .cold]
        ???:0
0x135181c ana::(anonymous
namespace)::fd_state_machine::on_stmt(ana::sm_context*, ana::supernode const*,
gimple const*) const
        ???:0
0x12fcd74 ana::exploded_node::on_stmt(ana::exploded_graph&, ana::supernode
const*, gimple const*, ana::program_state*, ana::uncertainty_t*,
ana::path_context*)
        ???:0
0x12ff7ab ana::exploded_graph::process_node(ana::exploded_node*)
        ???:0
0x130064a ana::exploded_graph::process_worklist()
        ???:0
0x1302999 ana::impl_run_checkers(ana::logger*)
        ???:0
0x130394e ana::run_checkers()
        ???:0
0x12f37e8 (anonymous namespace)::pass_analyzer::execute(function*)
        ???:0

$ gcc -v |& unnix
Using built-in specs.
COLLECT_GCC=/<<NIX>>/gcc-13.0.0/bin/gcc
COLLECT_LTO_WRAPPER=/<<NIX>>/gcc-13.0.0/libexec/gcc/x86_64-unknown-linux-gnu/13.0.0/lto-wrapper
Target: x86_64-unknown-linux-gnu
Configured with:
Thread model: posix
Supported LTO compression algorithms: zlib
gcc version 13.0.0 20220807 (experimental) (GCC)

[Bug analyzer/106551] [13 Regression] dup2 causes -fanalyser ICE in valid_to_unchecked_state, at analyzer/sm-fd.cc:751

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=106551

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |13.0

[Bug analyzer/106551] [13 Regression] dup2 causes -fanalyser ICE in valid_to_unchecked_state, at analyzer/sm-fd.cc:751

[dmalcolm at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=106551

David Malcolm <dmalcolm at gcc dot gnu.org> changed:

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

--- Comment #1 from David Malcolm <dmalcolm at gcc dot gnu.org> ---
Thanks for filing this; almost certainly related to 
https://gcc.gnu.org/git/?p=gcc.git;a=commitdiff;h=6a11f2d974a912aaaedb0ce32cdfde10193003cd

Immad, do you want to take a look?

[Bug analyzer/106551] [13 Regression] dup2 causes -fanalyzer ICE in valid_to_unchecked_state, at analyzer/sm-fd.cc:751

[mir at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=106551

--- Comment #2 from Immad Mir <mir at gcc dot gnu.org> ---
 Sergei Trofimovich: Thanks for bringing the issue to our attention.

Dave: I've sent a patch via gcc-patches.

[Bug analyzer/106551] [13 Regression] dup2 causes -fanalyzer ICE in valid_to_unchecked_state, at analyzer/sm-fd.cc:751

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=106551

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

https://gcc.gnu.org/g:837142257cbde3cc03ee0dacd1d7b2fb4fa48bae

commit r13-2023-g837142257cbde3cc03ee0dacd1d7b2fb4fa48bae
Author: Immad Mir <mirimmad@outlook.com>
Date:   Thu Aug 11 21:45:54 2022 +0530

    analyzer: fix ICE casued by dup2 in sm-fd.cc[PR106551]

    This patch fixes the ICE caused by valid_to_unchecked_state,
    at analyzer/sm-fd.cc by handling the m_start state in
    check_for_dup.

    Tested lightly on x86_64.

    gcc/analyzer/ChangeLog:
            PR analyzer/106551
            * sm-fd.cc (check_for_dup): handle the m_start
            state when transitioning the state of LHS
            of dup, dup2 and dup3 call.

    gcc/testsuite/ChangeLog:
            * gcc.dg/analyzer/fd-dup-1.c: New testcases.
            * gcc.dg/analyzer/fd-uninit-1.c: Remove bogus
            warning.
    Signed-off-by: Immad Mir <mirimmad@outlook.com>

[Bug analyzer/106551] [13 Regression] dup2 causes -fanalyzer ICE in valid_to_unchecked_state, at analyzer/sm-fd.cc:751

[tlange at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=106551

Tim Lange <tlange at gcc dot gnu.org> changed:

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

--- Comment #4 from Tim Lange <tlange at gcc dot gnu.org> ---
Hi Immad,

while testing my patch on coreutils [1], I encountered another ICE at the same
line. I'm using the newest commit on the master at the time of writing this [2]
(I've double-checked that I indeed have your fix included).

during IPA pass: analyzer
../lib/freopen.c: In function 'rpl_freopen':
../lib/freopen.c:82:19: internal compiler error: in valid_to_unchecked_state,
at analyzer/sm-fd.cc:751
   82 |               if (dup2 (nullfd, fd) < 0)
      |                   ^~~~~~~~~~~~~~~~~
0x7bc106 valid_to_unchecked_state
        ../../gcc/analyzer/sm-fd.cc:751
0x7bc106 valid_to_unchecked_state
        ../../gcc/analyzer/sm-fd.cc:742
0x7bc106 check_for_dup
        ../../gcc/analyzer/sm-fd.cc:990
0x12a5e8c on_stmt
        ../../gcc/analyzer/sm-fd.cc:805
0x124f4e4 ana::exploded_node::on_stmt(ana::exploded_graph&, ana::supernode
const*, gimple const*, ana::program_state*, ana::uncertainty_t*,
ana::path_context*)
        ../../gcc/analyzer/engine.cc:1449
0x1251f1b ana::exploded_graph::process_node(ana::exploded_node*)
        ../../gcc/analyzer/engine.cc:3868
0x1252dba ana::exploded_graph::process_worklist()
        ../../gcc/analyzer/engine.cc:3271
0x125514d ana::impl_run_checkers(ana::logger*)
        ../../gcc/analyzer/engine.cc:5912
0x125614e ana::run_checkers()
        ../../gcc/analyzer/engine.cc:5986
0x1245ee8 execute
        ../../gcc/analyzer/analyzer-pass.cc:87


[1] compiling coreutils commit:
https://git.savannah.gnu.org/gitweb/?p=coreutils.git;a=commit;h=9f71f478ec6929d323c17f0482db8791de87b4fd
[2]
https://gcc.gnu.org/git/?p=gcc.git;a=commit;h=fe4dc4f553d8639b0ce3465e5e6dcf61c1ae9752

[Bug analyzer/106551] [13 Regression] dup2 causes -fanalyzer ICE in valid_to_unchecked_state, at analyzer/sm-fd.cc:751

[mir at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=106551

--- Comment #5 from Immad Mir <mir at gcc dot gnu.org> ---
Thanks for reporting the ICE, Tim.

Dave: I have sent a patch via gcc-patches.

[Bug analyzer/106551] [13 Regression] dup2 causes -fanalyzer ICE in valid_to_unchecked_state, at analyzer/sm-fd.cc:751

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=106551

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

https://gcc.gnu.org/g:ed7e7620188dc0e653815bdd40d8802bd5446638

commit r13-2051-ged7e7620188dc0e653815bdd40d8802bd5446638
Author: Immad Mir <mirimmad@outlook.com>
Date:   Mon Aug 15 22:21:27 2022 +0530

    analyzer: fix for ICE in sm-fd.cc [PR106551]

    This patch fixes the ICE caused by valid_to_unchecked_state
    in sm-fd.cc by exiting early if first argument of any "dup"
    functions is invalid.

    gcc/analyzer/ChangeLog:
            PR analyzer/106551
            * sm-fd.cc (check_for_dup): exit early if first
            argument is invalid for all dup functions.

    gcc/testsuite/ChangeLog:
            PR analyzer/106551
            * gcc.dg/analyzer/fd-dup-1.c: New testcase.

    Signed-off-by: Immad Mir <mirimmad@outlook.com>

[Bug analyzer/106551] [13 Regression] dup2 causes -fanalyzer ICE in valid_to_unchecked_state, at analyzer/sm-fd.cc:751

[mir at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=106551

--- Comment #7 from Immad Mir <mir at gcc dot gnu.org> ---
(In reply to CVS Commits from comment #6)

The patch was lightly tested on x86_64 Linux.