February 2011 - Mpich2-core - mailman3.cels.anl.gov

mpich2-1.3.3rc1
by Pavan Balaji 03 Mar '11

03 Mar '11

I've created the tarball for 1.3.3rc1 and updated the staging website: http://www.mcs.anl.gov/research/projects/mpich2staging/balaji/mpich2/downlo… Please try out the tarball. If everything works fine, we can push out the release tomorrow. -- Pavan -- Pavan Balaji http://www.mcs.anl.gov/~balaji

3 5

1.3.x branch failing
by Rajeev Thakur 25 Feb '11

25 Feb '11

I am switching the nightly tests to use the 1.3.3 branch for the upcoming release. However, all tests failed last night in the 1.3.x branch that is currently being used. They worked the previous night. http://www.mcs.anl.gov/research/projects/mpich2/nightly/old/ Rajeev

4 6

Coming in a little late
by Pavan Balaji 18 Feb '11

18 Feb '11

I'm taking this morning off as half-a-day of vacation. I'll be coming in after lunch. -- Pavan -- Pavan Balaji http://www.mcs.anl.gov/~balaji

1 0

Urgent 1.3.2p1 release
by Pavan Balaji 14 Feb '11

14 Feb '11

All, I got a bug report on mpich2-1.3.2 about a failure with older GNU compilers. I've fixed it and committed it to the trunk, 1.3.x branch and 1.3.2 branch. I'd like to push out an urgent patch release (1.3.2p1) with this fix. If there are other critical fixes that can't wait till 1.3.3, please let me know and we can push them out as well. Rajeev: can you move the nightly tests to the 1.3.2 branch (not 1.3.x branch)? If tonight's nightly tests are clean, we can push it out tomorrow. Thanks, -- Pavan -- Pavan Balaji http://www.mcs.anl.gov/~balaji

3 4

make distclean results in hydra running configure checks
by Jayesh Krishna 10 Feb '11

10 Feb '11

Running "make distclean" (on trunk r7924) results in all these configure checks. =========================== snip - configure.log ============================== make[3]: Entering directory `/sandbox/jayesh/freshBuild/src/pm/hydra' CDPATH="${ZSH_VERSION+.}:" && cd . && /bin/sh /sandbox/jayesh/freshBuild/src/pm/hydra/confdb/missing --run aclocal-1.11 -I confdb cd . && /bin/sh /sandbox/jayesh/freshBuild/src/pm/hydra/confdb/missing --run automake-1.11 --foreign CDPATH="${ZSH_VERSION+.}:" && cd . && /bin/sh /sandbox/jayesh/freshBuild/src/pm/hydra/confdb/missing --run autoconf /bin/sh ./config.status --recheck running CONFIG_SHELL=/bin/sh /bin/sh /sandbox/jayesh/freshBuild/src/pm/hydra/configure --prefix=/sandbox/jayesh/freshBuild/mpich2-install --disable-f77 --disable-fc --disable-cxx --disable-mpe --disable-romio --disable-option-checking CC=gcc CFLAGS= -O2 LDFLAGS= LIBS=-lpthread CPPFLAGS= -I/sandbox/jayesh/freshBuild/src/mpl/include -I/sandbox/jayesh/freshBuild/src/mpl/include -I/sandbox/jayesh/freshBuild/src/openpa/src -I/sandbox/jayesh/freshBuild/src/openpa/src -I/sandbox/jayesh/freshBuild/src/mpid/ch3/include -I/sandbox/jayesh/freshBuild/src/mpid/ch3/include -I/sandbox/jayesh/freshBuild/src/mpid/common/datatype -I/sandbox/jayesh/freshBuild/src/mpid/common/datatype -I/sandbox/jayesh/freshBuild/src/mpid/common/sched -I/sandbox/jayesh/freshBuild/src/mpid/common/sched -I/sandbox/jayesh/freshBuild/src/mpid/common/locks -I/sandbox/jayesh/freshBuild/src/mpid/common/locks -I/sandbox/jayesh/freshBuild/src/mpid/ch3/channels/nemesis/include -I/sandbox/jayesh/freshBuild/src/mpid/ch3/channels/nemesis/include -I/sandbox/jayesh/freshBuild/src/mpid/ch3/channels/nemesis/nemesis/include -I/sandbox/jayesh/freshBuild/src/mpid/ch3/channels/nemesis/nemesis/include -I/sandbox/jayesh/freshBuild/src/mpid/ch3/channels/nemesis/nemesis/utils/monitor -I/sandbox/jayesh/freshBuild/src/mpid/ch3/channels/nemesis/nemesis/utils/monitor -I/sandbox/jayesh/freshBuild/src/util/wrappers -I/sandbox/jayesh/freshBuild/src/util/wrappers --no-create --no-recursion checking for a BSD-compatible install... /usr/bin/install -c checking whether build environment is sane... yes checking for a thread-safe mkdir -p... /bin/mkdir -p checking for gawk... gawk checking whether make sets $(MAKE)... yes checking for style of include used by make... GNU checking whether the C compiler works... yes checking for C compiler default output file name... a.out checking for suffix of executables... checking whether we are cross compiling... no checking for suffix of object files... o ... =========================== snip - configure.log ============================== -Jayesh

2 2

MPICH2 on POWER7
by William Gropp 09 Feb '11

09 Feb '11

I'd like to make some introductions between the Illinois/NCSA group working on MPICH2 on POWER7 and the ANL group. Kitrick Sheets - Is working on shared memory performance in Nemesis for Power7 Gopalakrishnan Santhanaraman - Is working on RMA enhancements, particularly the hybrid lazy/eager approach for passive target Kalyana Chadalavada - Is working on MPI-3 updates, particularly the RMA chapter (We are also collecting baseline performance data, on both the GA product and the upcoming system that is what will go into Blue Waters.) Tarun Prabhu and Esteban Rojas have already been introduced; they continue to work on the LAPI (and eventually PAMI) port and will be looking at Datatype enhancements as well, as Torsten has found an application where datatypes provide a significant benefit. Please welcome Kitrick, Gopal, and Kalyana and keep them in mind when looking at related parts of the code. Thanks! Bill William Gropp Deputy Director for Research Institute for Advanced Computing Applications and Technologies Paul and Cynthia Saylor Professor of Computer Science University of Illinois Urbana-Champaign

6 5

Planning for the mpich2-1.3.3 release
by Pavan Balaji 03 Feb '11

03 Feb '11

All, The next bug fix release of MPICH2 (1.3.3) is planned for Apr. 15th. Please see the tickets assigned to you over here: https://trac.mcs.anl.gov/projects/mpich2/report/11 Please start working on your tickets early, and do not wait till we get close to the release deadline. Thanks, -- Pavan -- Pavan Balaji http://www.mcs.anl.gov/~balaji

4 6