GitLab

In the MPIC_Sendrecv functions, the status object should always be defined
since we use it internally. This won't have any impact on performance since
the default is always to have FT collectives turned on anyway, but it will
prevent a crash when someone overwrites that default.

Fixes #2026
Signed-off-by: Sangmin Seo <sseo@anl.gov>

Pavan Balaji authored Apr 12, 2014 and Kenneth Raffenetti committed Apr 15, 2014

All ROMIO public symbols are being added into the libromio convenience
library, which is LIBADD'ed into libmpi.  The correct behavior would
be for libtool to add -Wl,--whole-archive to the link line, thus
forcing all symbols in libromio to end up in libmpi.  However, in some
cases, this is not working out correctly.  For example, with the
Absoft compiler, the linker flags were being reordered thus causing
the -Wl,--whole-archive flag to not correspond to libromio.
Consequently, only symbols that were used from within the other
functions of libmpi were being included.

For example, Absoft's linker reordering is below:

----8<----
% af77 -v -shared -fPIC src/mpi/attr/.libs/lib_libmpi_la-attrutil.o src/mpi_t/.libs/lib_libmpi_la-pvar_write.o \
  -Wl,--whole-archive src/mpi/romio/.libs/libromio.a -Wl,--no-whole-archive  \
  -Wl,-rpath -Wl,/sandbox/balaji/build-absoft/src/mpl/.libs \
  -Wl,-rpath -Wl,/sandbox/balaji/build-absoft/src/openpa/src/.libs \
  -Wl,-rpath -Wl,/sandbox/balaji/build-absoft/install/lib \
  -L/sandbox/balaji/build-absoft/src/mpl -L/sandbox/balaji/build-absoft/src/openpa/src \
  /sandbox/balaji/build-absoft/src/mpl/.libs/libmpl.so /sandbox/balaji/build-absoft/src/openpa/src/.libs/libopa.so \
  -lrt -lpthread  -O0   -Wl,-soname -Wl,libmpi.so.12 -o lib/.libs/libmpi.so.12.0.0

gcc src/mpi/attr/.libs/lib_libmpi_la-attrutil.o src/mpi_t/.libs/lib_libmpi_la-pvar_write.o \
    /sandbox/balaji/build-absoft/src/mpl/.libs/libmpl.so /sandbox/balaji/build-absoft/src/openpa/src/.libs/libopa.so \
    -L/soft/com/packages/absoft11.0/shlib64 -Wl,--whole-archive -Wl,--no-whole-archive \
    -Wl,-rpath -Wl,/sandbox/balaji/build-absoft/src/mpl/.libs \
    -Wl,-rpath -Wl,/sandbox/balaji/build-absoft/src/openpa/src/.libs \
    -Wl,-rpath -Wl,/sandbox/balaji/build-absoft/install/lib \
    -L/sandbox/balaji/build-absoft/src/mpl -L/sandbox/balaji/build-absoft/src/openpa/src \
    -Wl,-soname -Wl,libmpi.so.12 -shared -o lib/.libs/libmpi.so.12.0.0 src/mpi/romio/.libs/libromio.a \
    -lrt -lpthread -laf90math -lafio -lamisc -laf77math -lm -lmv
----8<----

In the past, we got lucky in two ways:

1. We were including all the Fortran 77 symbols into libmpi.  Since
the Fortran symbols were calling the C ROMIO symbols, those ROMIO
symbols were pulled in even without -Wl,--whole-archive.

2. The problem only showed up with absoft and nag, both of which are
Fortran-only compilers.  If a C compiler has this issue, it should
have shown up when Fortran support is disabled.
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

Pavan Balaji authored Apr 15, 2014 and Kenneth Raffenetti committed Apr 15, 2014

We were adding both the ROMIO MPI_ and PMPI_ symbols into the
external_libs variable, which end up in the same library.  This patch
separates them out.  It also cleans up the variables to which other
libtool libraries (such as the debug library) are added.
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

Pavan Balaji authored Apr 15, 2014 and Kenneth Raffenetti committed Apr 15, 2014

We were not explicitly linking the application binaries to libpmpi
earlier.  This works fine when the application uses the MPI_ symbols,
but not when the application uses the PMPI_ symbols.  On mac, where a
separate profiling library is built, we were using -flat_namespace to
supress this issue.  But on other platforms, without this flag, the
issue showed up when passing --disable-weak-symbols.
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

MPIR_ERRTEST_ARGNULL is incorrect for validating an MPI_Info
handle, since MPI_INFO_NULL is not defined as 0. This updates the
test in MPI_Comm_set_info to match the one in MPI_Win_set_info.

Patch originally from thakur@mcs.anl.gov

if GPFS enabled, we were declaring all file systems GPFS.  This holdover
from the old BlueGene-specific days does not belong here.
Signed-off-by: Paul Coffman <pkcoff@us.ibm.com>

20 years of portability workarounds make for quite a tangle.  Only one
of the four methods is ever used in practice now.  Provide comments
about the older approaches and indicate which one we expect to be the
common case.
Signed-off-by: Paul Coffman <pkcoff@us.ibm.com>

only ad_fstype.c uses GPFS_SUPER_MAGIC now.  Also, GPFS_SUPER_MAGIC is
properly provided by gpfs.h, but we can fall back to old way of
declaring it ourselves if gpfs.h is not around.
Signed-off-by: Paul Coffman <pkcoff@us.ibm.com>

Signed-off-by: Pavan Balaji <balaji@mcs.anl.gov>

Fixes compiler warnings in ch3u_rma_ops.c and ch3u_rma_acc_ops.c. Complains
were about "orig_vc" and "target_vc" of potentially being unused. This is a
false positive, as the conditional statements were correct. Warning fixed by just
initializing the pointers to NULL. gcc 4.6.3 complained in stomp, whereas gcc
4.7.4 did not on my laptop.
Signed-off-by: Pavan Balaji <balaji@mcs.anl.gov>

Those were related to pragma weak alias without previous external definitions.
Signed-off-by: Pavan Balaji <balaji@mcs.anl.gov>

Fixes complain by icc
Signed-off-by: Pavan Balaji <balaji@mcs.anl.gov>

Complained by icc
Signed-off-by: Pavan Balaji <balaji@mcs.anl.gov>

The variable is declared as external in pmip.h, but defined in three
different files. The linker warns that it's actually merging them. If we
want to use an only variable declared as 'extern', it should be only
defined in one place. This commit removes two out of the three
declarations. See #1966.
Signed-off-by: Pavan Balaji <balaji@mcs.anl.gov>

Fixes the following warning when --enable-fast is enabled:

src/mpi/coll/bcast.c:1012:23: warning: suggest braces around empty body
in an 'if' statement [-Wempty-body]

See #1966.
Signed-off-by: Pavan Balaji <balaji@mcs.anl.gov>

This adds the same checks as the previous commit to the fortran tests. For the
most part, the F77 tests are copied over to the F90 tests so little was added
to the F90 directory.

Fixes #1699
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

The test suite assumes that it is running on a system where
MPI_COMM_SPAWN(_MULTIPLE) is possible. This isn't true on all systems and thus
prevents the test suite from being useful everywhere.

This check (contributed by Jeff H. - jhammond@alcf.anl.gov) looks to see if
the MPI_COMM_WORLD attribute MPI_UNIVERSE_SIZE is available and actually
greater than the size of MPI_COMM_WORLD.

If the spawn tests will fail because spawning is not possible on the system
(whether because the universe size is too small or spawning is not supported),
the tests will return success (unless there was some error in the detection
process).
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

Signed-off-by: Wesley Bland <wbland@anl.gov>

Add a test to check for correct behavior of MPI_Accumulate operations
with the MPI_MINLOC and MPI_MAXLOC operations.
Signed-off-by: Xin Zhao <xinzhao3@illinois.edu>

libmpi depends on libpmpi.  We need to build them in that order to
make sure the dependencies are satisfied.  libtool should do this
automatically, but it does not.  More description of the error is
available here: https://bugzilla.redhat.com/show_bug.cgi?id=91110

Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

if a file system supports scalable resize, we can continue to defer
open.  I had the logic backwards and conduced the deferred open when we
did not need to (and did not when we did).  Thanks, Paul Coffman.
Signed-off-by: Paul Coffman <pkcoff@us.ibm.com>

Paul Coffman authored Apr 01, 2014 and Rob Latham committed Apr 02, 2014

in the final sort on the bridgelist,  was starting an index at -1
instead of 0, resulting in typically bizzare (occaional) memory errors.
Signed-off-by: Rob Latham <robl@mcs.anl.gov>

No need for this Makefile.mk to add BGL_OPTIM flags to CFLAGS: nothing
in the code uses them.
Signed-off-by: Paul Coffman <pkcoff@us.ibm.com>

The default optimization flags (-O2) are set for C and F77, but not for
C++ and Fortran 90.

In commit [b4ce3c99

], the behaviour of PAC_C_CHECK_COMPILER_OPTION and
PAC_F77_CHECK_COMPILER_OPTION were changed in configure.ac, while
PAC_CXX_CHECK_COMPILER_OPTION and PAC_FC_CHECK_COMPILER_OPTION
remained the same.

Fixes #2033
Signed-off-by: Pavan Balaji <balaji@mcs.anl.gov>

Signed-off-by: Huiwei Lu <huiweilu@mcs.anl.gov>

Once the process mapping is complete, we need to make another round of
node ID updates for the case where MPIR_CVAR_CH3_ODD_EVEN_CLIQUES is
set.  We were not doing that, so the CVAR variable had no effect.
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

Create symlinks for libmpich and friends, for applications that might
still be relying on those names.
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

No reviewer.

Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

When the conditions are right, we use libtool's interlibrary
dependencies to only link the application with the minimum set of
libraries required.  Those libraries will pull in the other required
libraries on demand.  That way, the application library dependencies
would be minimal, allowing us to add/remove extra libraries in the
future.

The "conditions are not right" when one of the following is true:

1. MPICH shared library builds are disabled.

2. libtool doesn't support interlibrary dependencies on this platform.

3. The user passed -static or -mpich-alllibs to the command line.

4. Interlibrary dependency tracking is explicitly disabled at
configure time.

Fixes #2040.
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

Pavan Balaji authored Mar 31, 2014 and Kenneth Raffenetti committed Apr 01, 2014

We moved the symbols around in libmpi and libpmpi.  pkg-config should
match this.
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

Pavan Balaji authored Mar 31, 2014 and Kenneth Raffenetti committed Apr 01, 2014

Maintain a list of files that go into each library.  If a particular
binding is not enabled, the list variable still exists, but will just
be empty.  This simplifies the management of which files/symbols go
into which library.

Move all MPI_ symbols to the libmpi library and all other symbols to
the libpmpi library.  All Fortran 77 symbols go into libmpif77.so,
while C symbols go into libmpi.so.  There are some exceptions, such as
status_f2c, which are handled by the Fortran code but used in C.  Our
Fortran 90 build only creates a few symbols and uses the f77 symbols
for everything else.  These few symbols go into libmpifort.so.

Also update compiler wrappers to link to correct libraries.  mpif77
should now link with libmpif77.  mpif90 links with both libmpifort and
libmpif77, since our F90 build still keeps the core Fortran library
symbols in libf77.

We completely ignored the F77 library earlier.  This was OK because
all of the Fortran symbols were ending up in libmpi.  Now that we have
separated out the symbols to the right library, we now need to link to
libmpif77 as well.

Also added inter-library dependencies.

libmpi has a dependency on several internal libraries: libmpl, libopa.
libmpicxx did not have a dependency on libmpi, added.
libmpif77 did not have a dependency on libmpi, added.
libmpifort did not have a dependency on libmpi, added.

This dependency model is sufficient for C and F77, but not for C++ and
F90.  The C and F77 libraries contain all the symbols the application
relies on, but the F90 and C++ libraries don't.  In the case of F90,
symbols such as mpi_bcast are missing and are borrowed from the F77
library.  In the case of C++, mpicxx.h contains calls directly to C
functions (such as MPI_Reduce_local), which get embedded into the
application.

Fixes #2023.
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

Pavan Balaji authored Mar 29, 2014 and Kenneth Raffenetti committed Apr 01, 2014

Rename BUILD_CXX_LIB to BUILD_CXX_BINDING.  Rename BUILD_F90_LIB to
BUILD_FC_BINDING.
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

Pavan Balaji authored Mar 31, 2014 and Kenneth Raffenetti committed Apr 01, 2014

Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

Pavan Balaji authored Mar 23, 2014 and Kenneth Raffenetti committed Apr 01, 2014

The following library names are used to make the naming consistent
across the ABI compatibility group:

C libraries: libmpi.* and libpmpi.*
C++ library: libmpicxx.*
F77 libraries: libmpif77.*
F90+ library: libmpifort.*

This patch also gets rid of the FWRAPNAME variable, which is a
duplicate of MPIFLIBNAME.  Similarly, FCWRAPNAME is removed and a new
variable MPIFCLIBNAME is added, so it's consistent with the other
names.

PMPIFLIBNAME, which was unused, is no longer present.

Fixes #2039.
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

Pavan Balaji authored Mar 28, 2014 and Kenneth Raffenetti committed Apr 01, 2014

This symbol should go into libmpich, not libpmpich.
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

No Reviewer

No reviewer.