- remove dead code that isn't used anywhere (originally ompi_fifo_t
was going to be a generalized class, but now it's exclusively used
in the sm stuff, so there's no point in the generalized code that
definitely *won't* work with the sm btl, or is not being used now
[SVN always has history so we can go back])
- had to add an interprocess lock in the area where the writer may
create a new circular buffer to ensure that the reader's tail
doesn't accidentally end up back in the same old buffer while the
head continues on to a new circular buffer (this was what was
happening to cause some intel tests to hang -- e.g., MPI_Scan_c,
MPI_Send_fairness_c and MPI_Isend_fairness_c). Unbelievably, this
may actually *increase* performance because it may order things
better. Will do performance testing tomorrow. We're fairly certain
that this lock can probably be removed and the code fixed in a
different way, but we're under a deadline and correctness comes
first, so it's been added to the to-do list to come back and
re-examine this case later.
This commit was SVN r8136.
- Add big comment about a general overview of what the sm btl is doing
- random small code cleanups
- fix instances of mca_btl_sm[0] to mca_btl_sm[1] where relevant
- remove a lot of unused, confusing, and incorrect interface functions
from ompi_fifo.h and ompi_circular_buffer.h. These functions, if
they were used, would not work properly with the scheme that the sm
btl uses with the fifos (i.e., receiver makes right -- if necessary)
- add some missing offset computations in the fifo and circular buffers
- change the types of offsets to be ssize_t, not size_t
- remove an offset parameter from a function that didn't need it
This commit was SVN r8135.
- Ensure to convert base_shared_mem_flags to be a relative offset in
the global storage, and then to convert that back to an absolute
virtual address before we try to use it
- Don't double increment n_local_procs when calculating the peer rank
during bootstrapping of the different base address case
Something else is still wrong; if mmap() returns a different base
address, things don't work (i.e., segv or hang forever when you try to
send a message). More specifically, the bootstrapping now seems to
correctly handle the case when mmap() base addresses are different,
but the message passing does *not* -- it always assumes that the
mmap() base addresses are the same.
Still working on the fix for that -- want to checkpoint what has been
done so far to facilitate working on different machines...
This commit was SVN r8134.
Lots of misc fixes: printfs->opal_output, handles fanin/out correctly for forced ops
unused vars, correct calculations on meaning of 'msgsize' for decision functions
(varies depending on algorithm), etc
This commit was SVN r8113.
When compiling C++ code that includes something that looks for the C++
header file "memory" (stupid C++ headers not having .h extensions), it
goes through the header file search path, which includes $(topsrcdir)/opal,
so it finds the directory $(topsrcdir)/opal/memory/ and tries to load
that as the memory header file and all goes downhill.
This commit was SVN r8111.
spining in orte_iof_base_flush() when running
intel_tests/src/MPI_Errhandler_fatal_c
When we close an endpoint by taking it out of the envent handler, we need to make
sure that it fits the criteria to pass through orte_iof_base_flush(), specificly
make sure we clean out the ep_frags list.
Note: This is more of a sanity check, since the endpoint should already be
in this state at the point of closure.
Secondly in orte_iof_base_endpoint_read_handler(), if we determine that it is
necessary to close the endpoint we have to "return" after doing so, otherwise
we add another frag to the endpoint which will cause it to hang in
orte_iof_base_flush().
Bug go squish!
This commit was SVN r8109.
it doesn't support it -- the compiler will automatically convert the
unsupported type to a type that it *does* support. For example, if
you try to use INTEGER*16 and the compiler doesn't support it, it may
well automatically convert it to INTEGER*8 for you (!). So we have to
check the actual size of the type once we determine that the compiler
doesn't error if we try to use it (i.e,. the compiler *might* support
that type). If the size doesn't match the expected size, then the
compiler doesn't really support it.
The F77 configure code actually handled this properly. The F90 code
did not quite do it right. This patch brings the F90 code up to the
same structure as the F77 code, albiet not m4-ized properly. I also
added a comment to config/f77_check.m4 that explains *why* we do this
extra size check (because no explanation was given).
The impetus for this was that xlf* on OS X 10.3 was not recognizing
that INTEGER*16 was not supported, and mpi-f90-interfaces.h was being
assembled incorrectly. This patch fixes this problem.
There is still one more problem, but waiting for some help from Craig
R on that (function pointers in F90 declarations).
This commit was SVN r8107.
functions (e.g., MPI_REDUCE). We don't generate the back-end
subroutines for them (because it makes an expontential number of
subroutines, and compilers literally will segv), so we shouldn't
generate the f90 interfaces for them, either. This allows user's MPI
F90 apps to automaitcally fall through to the F77 bindings for these
functions.
This commit was SVN r8094.
OS X (you get an undefined symbol opal_event_lock). Since the code is
all #if 0'ed out, #if 0 out the header for now as well.
I believe console and openmpi are to be removed from OMPI before 1.0
release, so this doesn't need to go to the 1.0 branch
This commit was SVN r8089.
name was changed to shorten it too early (and then not restored), so
the "interface" name was not output correctly into
mpi-f90-interfaces.h. Change to make it like the other long functions
-- temporarily change it to a shorter name while outputing the
subroutines, and then revert it when outputting the end interface.
This commit was SVN r8086.
test isn't really needed to make sure the malloc intercept code was working,
so just revert to not checking for now.
This should go to the 1.0 branch, as what is there now is causing issues.
This commit was SVN r8081.
- first we setup the connections in the begining with all the peers
- MX does not handle well the case where several peers make connections to the same
destination simultaneously.
So I change the order in which we connect. First we compute our rank in the array,
then in a round-robin fashion we setup connection starting with our left neighboard.
This commit was SVN r8075.
This flag is inherited by all datatypes create with unavailable datatypes. Basically, we let the user create the wrong datatype but we dont let him using it for any pt2pt communications or any pack/unpack.
This commit was SVN r8069.
comm_fill_rest there is no need for calling ompi_set_group_rank, since
we know already the rank of the process in the new comm. In case the
process was not part of the new communicator (rank = MPI_UNDEFINED)
calling this function caused a segfault on some platforms.
This commit was SVN r8060.
