[Bug c/109593] New: valgrind error doing build of latest gcc trunk

[Bug c/109593] New: valgrind error doing build of latest gcc trunk

[dcb314 at hotmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109593

            Bug ID: 109593
           Summary: valgrind error doing build of latest gcc trunk
           Product: gcc
           Version: 13.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
          Assignee: unassigned at gcc dot gnu.org
          Reporter: dcb314 at hotmail dot com
  Target Milestone: ---

I just tried a valgrind build of gcc trunk. Configure line is

../trunk/configure  \
        --disable-multilib \
        --disable-bootstrap \
        --enable-checking=valgrind \
        --enable-languages=c,c++

I get:

/home/dcb36/gcc/working/./gcc/xgcc -B/home/dcb36/gcc/working/./gcc/
-B/home/dcb36/gcc/results.20230422.valgrind/x86_64-pc-linux-gnu/bin/
-B/home/dcb36/gcc/results.20230422.valgrind/x86_64-pc-linux-gnu/lib/ -isystem
/home/dcb36/gcc/results.20230422.valgrind/x86_64-pc-linux-gnu/include -isystem
/home/dcb36/gcc/results.20230422.valgrind/x86_64-pc-linux-gnu/sys-include    -g
-O2 -O2  -g -O2 -DIN_GCC    -W -Wall -Wno-narrowing -Wwrite-strings -Wcast-qual
-Wstrict-prototypes -Wmissing-prototypes -Wold-style-definition  -isystem
./include  -fpic -mlong-double-80 -DUSE_ELF_SYMVER -fcf-protection -mshstk -g
-DIN_LIBGCC2 -fbuilding-libgcc -fno-stack-protector  -fpic -mlong-double-80
-DUSE_ELF_SYMVER -fcf-protection -mshstk -I. -I. -I../.././gcc
-I../../../trunk.year/libgcc -I../../../trunk.year/libgcc/.
-I../../../trunk.year/libgcc/../gcc -I../../../trunk.year/libgcc/../include
-I../../../trunk.year/libgcc/config/libbid -DENABLE_DECIMAL_BID_FORMAT
-DHAVE_CC_TLS  -DUSE_TLS  -o _bswapdi2_s.o -MT _bswapdi2_s.o -MD -MP -MF
_bswapdi2_s.dep -DSHARED -DL_bswapdi2 -c ../../../trunk.year/libgcc/libgcc2.c
==2810006== Conditional jump or move depends on uninitialised value(s)
==2810006==    at 0xB9FD91: real_identical(real_value const*, real_value
const*) (real.cc:1318)
==2810006==    by 0xEBFAD4: frange::operator==(frange const&) const
(value-range.cc:685)
==2810006==    by 0x18BCCC0: operator!= (value-range.h:101)
==2810006==    by 0x18BCCC0: operator!= (value-range.h:646)
==2810006==    by 0x18BCCC0: ranger_cache::propagate_cache(tree_node*)
(gimple-range-cache.cc:1097)

[Bug c/109593] valgrind error doing build of latest gcc trunk

[sjames at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109593

Sam James <sjames at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |aldyh at gcc dot gnu.org,
                   |                            |amacleod at redhat dot com,
                   |                            |sjames at gcc dot gnu.org

--- Comment #1 from Sam James <sjames at gcc dot gnu.org> ---
Could you try bisect to speed up things?

[Bug c/109593] valgrind error doing build of latest gcc trunk

[dcb314 at hotmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109593

--- Comment #2 from David Binderman <dcb314 at hotmail dot com> ---
(In reply to Sam James from comment #1)
> Could you try bisect to speed up things?

I think my last valgrind build, dated 20230311, with
git hash 5de5e977aa0b653e was ok.

That's a range of 416 commits. Attempting a bisect.

[Bug c/109593] valgrind error doing build of latest gcc trunk

[dcb314 at hotmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109593

--- Comment #3 from David Binderman <dcb314 at hotmail dot com> ---
(In reply to David Binderman from comment #2)
> That's a range of 416 commits. Attempting a bisect.

Git range seems to be e6b050da8a4513ab..cbddd574a78529b9, which is 52 commits.

[Bug c/109593] valgrind error doing build of latest gcc trunk

[dcb314 at hotmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109593

--- Comment #4 from David Binderman <dcb314 at hotmail dot com> ---
(In reply to David Binderman from comment #3)
> Git range seems to be e6b050da8a4513ab..cbddd574a78529b9, which is 52
> commits.

Now downto 603fc926fee69ab3..10e481b154c5fc63, which is 13 commits.

[Bug c/109593] valgrind error doing build of latest gcc trunk

[aldyh at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109593

--- Comment #5 from Aldy Hernandez <aldyh at gcc dot gnu.org> ---
Huh.  I'm gonna guess:

commit 10e481b154c5fc63e6ce4b449ce86cecb87a6015
Author: Aldy Hernandez <aldyh@redhat.com>
Date:   Thu Jan 26 04:46:54 2023 +0100

    Return true from operator== for two identical ranges containing NAN.


-      if (known_isnan () || src.known_isnan ())
-       return false;
-
       return (real_identical (&m_min, &src.m_min)
              && real_identical (&m_max, &src.m_max)
              && m_pos_nan == src.m_pos_nan

If either operand is a NAN, m_min or m_max have garbage.  Can you verify that's
the commit?  Either way, I can take a look at it on Monday.

[Bug c/109593] valgrind error doing build of latest gcc trunk

[dcb314 at hotmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109593

--- Comment #6 from David Binderman <dcb314 at hotmail dot com> ---
(In reply to Aldy Hernandez from comment #5)
>  Can you verify that's the commit?  

git bisect says:

10e481b154c5fc63e6ce4b449ce86cecb87a6015 is the first bad commit.

[Bug tree-optimization/109593] valgrind error doing build of latest gcc trunk

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109593

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |ice-on-valid-code
          Component|c                           |tree-optimization

--- Comment #7 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to David Binderman from comment #6)
> (In reply to Aldy Hernandez from comment #5)
> >  Can you verify that's the commit?  
> 
> git bisect says:
> 
> 10e481b154c5fc63e6ce4b449ce86cecb87a6015 is the first bad commit.

r14-32-g10e481b154c5fc63

[Bug tree-optimization/109593] [14 Regression] valgrind error doing build of latest gcc trunk

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109593

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |14.0
            Version|13.0                        |14.0
            Summary|valgrind error doing build  |[14 Regression] valgrind
                   |of latest gcc trunk         |error doing build of latest
                   |                            |gcc trunk

[Bug tree-optimization/109593] [14 Regression] Valgrind error doing build of latest gcc trunk since r14-32-g10e481b154c5fc63

[aldyh at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109593

--- Comment #8 from Aldy Hernandez <aldyh at gcc dot gnu.org> ---
Created attachment 54908
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=54908&action=edit
untested patch

Patch in testing.  Does this fix the problem on the reporter's side?

[Bug tree-optimization/109593] [14 Regression] Valgrind error doing build of latest gcc trunk since r14-32-g10e481b154c5fc63

[dcb314 at hotmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109593

--- Comment #9 from David Binderman <dcb314 at hotmail dot com> ---
(In reply to Aldy Hernandez from comment #8)
> Does this fix the problem on the reporter's side?

Looks good to me.

[Bug tree-optimization/109593] [14 Regression] Valgrind error doing build of latest gcc trunk since r14-32-g10e481b154c5fc63

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109593

--- Comment #10 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:5830953a446959c5d8d9b569ffcd3e3bb16a06dc

commit r14-174-g5830953a446959c5d8d9b569ffcd3e3bb16a06dc
Author: Aldy Hernandez <aldyh@redhat.com>
Date:   Sun Apr 23 08:04:55 2023 +0200

    Handle NANs in frange::operator== [PR109593]

    This patch...
            commit 10e481b154c5fc63e6ce4b449ce86cecb87a6015
            Return true from operator== for two identical ranges containing
NAN.

    removed the check for NANs, which caused us to read from m_min and
    m_max which are undefined for NANs.

    gcc/ChangeLog:

            PR tree-optimization/109593
            * value-range.cc (frange::operator==): Handle NANs.

[Bug tree-optimization/109593] [14 Regression] Valgrind error doing build of latest gcc trunk since r14-32-g10e481b154c5fc63

[aldyh at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109593

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

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

--- Comment #11 from Aldy Hernandez <aldyh at gcc dot gnu.org> ---
fixed