This fix has been pushed into the main MPICH repo. Ken On 01/11/2016 06:44 AM, VandeVondele Joost wrote:

I can confirm that Daniel's patch also fixes the issue I was seeing.

Thanks for the quick resolution.

------------------------------------------------------------------------ ------------------------------------------------------------------------ *From:* Jeff Hammond [[email protected]] *Sent:* Monday, January 11, 2016 6:51 AM *To:* [email protected] *Subject:* Re: [mpich-discuss] valgrind warning in MPIR_Process_status

To be explicit, the other thread is http://lists.mpich.org/pipermail/devel/2016-January/000677.html and it is indeed the same issue.

I pushed dibanez-helper_fns-valgrind to mpich-review just now. https://trac.mpich.org/projects/mpich/ticket/2328 was created for tracking purposes.

Jeff

On Sun, Jan 10, 2016 at 5:04 PM, Jeff Hammond <[email protected] <mailto:[email protected]>> wrote:

I think this is the same issue reported on the devel list yesterday. If it is, Daniel Ibanez provided a patch that I've merged into my (completely unofficial fork on github) and will push to the official MPICH review repo tonight.

If you want to try and/or verify Daniel'spatch, see github/jeffhammond/mpich. The branch you want should be obvious from the name (I can't remember what it is off hand).

Best,

Jeff

On Sunday, January 10, 2016, VandeVondele Joost <[email protected] <mailto:[email protected]>> wrote:

Hi,

we upgraded from mpich 3.1.2 to 3.2 and our nightly builds under valgrind show a new valgrind warning, which I suppose originates from the mpi library, rather than our code (given the stack trace below and the fact that they are absent with 3.1):

There are two slightly distinct traces (alltoall and alltoallv):

==30165== Conditional jump or move depends on uninitialised value(s) ==30165== at 0x6E00056: MPIR_Process_status (mpiimpl.h:4394) ==30165== by 0x6E00056: MPIC_Waitall (helper_fns.c:774) ==30165== by 0x6D5A853: MPIR_Alltoall_intra (alltoall.c:369) ==30165== by 0x6D5B63C: MPIR_Alltoall (alltoall.c:564) ==30165== by 0x6D5B6EF: MPIR_Alltoall_impl (alltoall.c:599) ==30165== by 0x6D5BF07: PMPI_Alltoall (alltoall.c:722) ==30165== by 0x6AFF0DC: PMPI_ALLTOALL (alltoallf.c:276) ==30165== by 0x1D04699: __message_passing_MOD_mp_alltoall_i22

==30201== Conditional jump or move depends on uninitialised value(s) ==30201== at 0x6E00056: MPIR_Process_status (mpiimpl.h:4394) ==30201== by 0x6E00056: MPIC_Waitall (helper_fns.c:774) ==30201== by 0x6D5C4EA: MPIR_Alltoallv_intra (alltoallv.c:194) ==30201== by 0x6D5CEC9: MPIR_Alltoallv (alltoallv.c:339) ==30201== by 0x6D5CFAA: MPIR_Alltoallv_impl (alltoallv.c:376) ==30201== by 0x6D5D8CA: PMPI_Alltoallv (alltoallv.c:527) ==30201== by 0x6AFF13E: PMPI_ALLTOALLV (alltoallvf.c:276) ==30201== by 0x1CCE180: __message_passing_MOD_mp_alltoall_z22v (message_passing_z.f90:223)

Is this issue already known/fixed ?

I saw there was a somewhat related change (which is part of 3.2):

https://trac.mpich.org/projects/mpich/changeset/be7c17b3177ccb0ede49761b44a4...

but maybe that change was not fully complete ?

Best regards,

Joost VandeVondele

------------------------------------------------------------------------ Prof. Dr. Joost VandeVondele Nanoscale Simulations

ETH Zurich HIT G 41.4 Wolfgang-Pauli-Strasse 27 8093 Zurich, Switzerland

[email protected] http://www.nanosim.mat.ethz.ch/ Tel.: +41 (0)44 6338451 ------------------------------------------------------------------------

-- Jeff Hammond [email protected] <mailto:[email protected]> http://jeffhammond.github.io/

-- Jeff Hammond [email protected] <mailto:[email protected]> http://jeffhammond.github.io/

_______________________________________________ discuss mailing list [email protected] To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss

_______________________________________________ discuss mailing list [email protected] To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss