[petsc-dev] Possible error running C/C++ src/snes/examples/tutorials/ex19 with 1 MPI process

[Antonio Trande]

On 10/10/2016 04:08 PM, Satish Balay wrote:
> Removing superlu_fixed the previous error? Thats wierd..

No; it was just an attempt.

> 
> You can try the attached patch - and see if that fixes this valgrind issue..
> 
> Satish
> 
> On Mon, 10 Oct 2016, Antonio Trande wrote:
> 
>> This is the only explicit error that i see (just on serial 32bit libs).
>> I have disabled SuperLU support, too.
>>
>> + cd petsc-3.7.4
>> + export
>> LD_LIBRARY_PATH=/usr/lib:/builddir/build/BUILD/petsc-3.7.4/petsc-3.7.4/i386/lib
>> +
>> LD_LIBRARY_PATH=/usr/lib:/builddir/build/BUILD/petsc-3.7.4/petsc-3.7.4/i386/lib
>> + export PETSC_DIR=/builddir/build/BUILD/petsc-3.7.4/petsc-3.7.4
>> + PETSC_DIR=/builddir/build/BUILD/petsc-3.7.4/petsc-3.7.4
>> + export PETSC_ARCH=i386
>> + PETSC_ARCH=i386
>> + export 'PETSCVALGRIND_OPTIONS= --tool=memcheck --leak-check=yes
>> --track-origins=yes'
>> + PETSCVALGRIND_OPTIONS=' --tool=memcheck --leak-check=yes
>> --track-origins=yes'
>> + make -C petsc-3.7.4 test
>> 'MPIEXEC=/builddir/build/BUILD/petsc-3.7.4/petsc-3.7.4/bin/petscmpiexec
>> -valgrind'
>> make: Entering directory '/builddir/build/BUILD/petsc-3.7.4/petsc-3.7.4'
>> Running test examples to verify correct installation
>> Using PETSC_DIR=/builddir/build/BUILD/petsc-3.7.4/petsc-3.7.4 and
>> PETSC_ARCH=i386
>> C/C++ example src/snes/examples/tutorials/ex19 run successfully with 1
>> MPI process
>> Possible error running C/C++ src/snes/examples/tutorials/ex19 with 2 MPI
>> processes
>> See http://www.mcs.anl.gov/petsc/documentation/faq.html
>> Uniprocessor version of MPI can only use one processor
>> Possible error running Fortran example src/snes/examples/tutorials/ex5f
>> with 1 MPI process
>> See http://www.mcs.anl.gov/petsc/documentation/faq.html
>> ==1577== Conditional jump or move depends on uninitialised value(s)
>> ==1577==    at 0x40323ED: strncpy (vg_replace_strmem.c:545)
>> ==1577==    by 0x47ECBE7: PetscStrncpy (str.c:414)
>> ==1577==    by 0x481C56E: PetscSetProgramName (options.c:314)
>> ==1577==    by 0x484BB5D: petscinitialize_ (zstart.c:306)
>> ==1577==    by 0x8049344: MAIN__ (ex5f.F:72)
>> ==1577==    by 0x8049808: main (ex5f.F:184)
>> ==1577==  Uninitialised value was created by a stack allocation
>> ==1577==    at 0x484BAC1: petscinitialize_ (zstart.c:270)
>> ==1577==
>> ==1577== Conditional jump or move depends on uninitialised value(s)
>> ==1577==    at 0x40323ED: strncpy (vg_replace_strmem.c:545)
>> ==1577==    by 0x47ECBE7: PetscStrncpy (str.c:414)
>> ==1577==    by 0x481C2FA: PetscGetProgramName (options.c:303)
>> ==1577==    by 0x47B0745: PetscErrorPrintfInitialize (errtrace.c:71)
>> ==1577==    by 0x484BCEF: petscinitialize_ (zstart.c:342)
>> ==1577==    by 0x8049344: MAIN__ (ex5f.F:72)
>> ==1577==    by 0x8049808: main (ex5f.F:184)
>> ==1577==  Uninitialised value was created by a stack allocation
>> ==1577==    at 0x484BAC1: petscinitialize_ (zstart.c:270)
>> ==1577==
>> Number of SNES iterations =     4
>> Completed test examples
>> =========================================
>>
>> Full log:
>> https://copr-be.cloud.fedoraproject.org/results/sagitter/petsc/fedora-rawhide-i386/00463045-petsc/build.log.gz
>>
>>

This error is disappeared.

Build on RHEL6 systems:
https://copr-be.cloud.fedoraproject.org/results/sagitter/petsc/epel-6-i386/00463135-petsc/build.log.gz

Build on Fedora 26 (development branch) systems:
https://copr-be.cloud.fedoraproject.org/results/sagitter/petsc/fedora-rawhide-i386/00463135-petsc/build.log.gz

What does mean the line

+ make -C petsc-3.7.4 test
'MPIEXEC=/builddir/build/BUILD/petsc-3.7.4/petsc-3.7.4/bin/petscmpiexec
-valgrind'
make: Entering directory '/builddir/build/BUILD/petsc-3.7.4/petsc-3.7.4'
Running test examples to verify correct installation
Using PETSC_DIR=/builddir/build/BUILD/petsc-3.7.4/petsc-3.7.4 and
PETSC_ARCH=i386
C/C++ example src/snes/examples/tutorials/ex19 run successfully with 1
MPI process
*Possible error running C/C++ src/snes/examples/tutorials/ex19 with 2
MPI processes*
See http://www.mcs.anl.gov/petsc/documentation/faq.html
Uniprocessor version of MPI can only use one processor
Fortran example src/snes/examples/tutorials/ex5f run successfully with 1
MPI process
Completed test examples

?



-- 
---
Antonio Trande
mailto: sagitter 'at' fedoraproject 'dot' org
http://fedoraos.wordpress.com/
https://fedoraproject.org/wiki/User:Sagitter
GPG Key: 0x6CE6D08A
Check on https://keys.fedoraproject.org/

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 473 bytes
Desc: OpenPGP digital signature
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20161010/c94591ee/attachment.sig>