and https://svn.open-mpi.org/trac/ompi/ticket/1853, mallopt() hints do
not always work -- it is possible for memory to be returned to the OS
and therefore OMPI's registration cache becomes invalid.
This commit removes all use of mallopt() and uses a different way to
integrate ptmalloc2 than we have done in the past. In particular, we
use almost exactly the same technique as MX:
* Remove all uses of mallopt, to include the opal/memory mallopt
component.
* Name-shift all of OMPI's internal ptmalloc2 public symbols (e.g.,
malloc -> opal_memory_ptmalloc2_malloc).
* At run-time, use the existing glibc allocator malloc hook function
pointers to fully hijack the glibc allocator with our own
name-shifted ptmalloc2.
* Make the decision whether to hijack the glibc allocator ''at run
time'' (vs. at link time, as previous ptmalloc2 integration
attempts have done). Look at the OMPI_MCA_mpi_leave_pinned
and OMPI_MCA_mpi_leave_pinned_pipeline environment variables and
the existence of /sys/class/infiniband to determine if we should
install the hooks or not.
* As an added bonus, we can now tell if libopen-pal is linked
statically or dynamically, and if we're linked statically, we
assume that munmap intercept support doesn't work.
See the opal/mca/memory/ptmalloc2/README-open-mpi.txt file for all the
gory details about the implementation.
Fixes trac:1853.
This commit was SVN r20921.
The following Trac tickets were found above:
Ticket 1853 --> https://svn.open-mpi.org/trac/ompi/ticket/1853
In case we use memcmp, strlen, strup and friends include <string.h>
Also several constants.h are not included directly
- Let's have mca_topo_base_cart_create return ompi-errors in
ompi/mca/topo/base/topo_base_cart_create.c
This commit was SVN r20773.
* add "register" function to mca_base_component_t
* converted coll:basic and paffinity:linux and paffinity:solaris to
use this function
* we'll convert the rest over time (I'll file a ticket once all
this is committed)
* add 32 bytes of "reserved" space to the end of mca_base_component_t
and mca_base_component_data_2_0_0_t to make future upgrades
[slightly] easier
* new mca_base_component_t size: 196 bytes
* new mca_base_component_data_2_0_0_t size: 36 bytes
* MCA base version bumped to v2.0
* '''We now refuse to load components that are not MCA v2.0.x'''
* all MCA frameworks versions bumped to v2.0
* be a little more explicit about version numbers in the MCA base
* add big comment in mca.h about versioning philosophy
This commit was SVN r19073.
The following Trac tickets were found above:
Ticket 1392 --> https://svn.open-mpi.org/trac/ompi/ticket/1392
a standalone library named libopenmpi-malloc. Users wanting to
use leave_pinned with ptmalloc2 will now need to link the library
into their application explicitly. All other users will use the
libc-provided allocator instead of Open MPI's ptmalloc2. This change
may be overriden with the configure option enable-ptmalloc2-internal
- The leave_pinned options will now default to using mallopt on
Linux in the cases where ptmalloc2 was not linked in. mallopt
will also only be available if munmap can be intercepted (the
default whenever Open MPI is not compiled with --without-memory-
manager.
- Open MPI will now complain and refuse to use leave_pinned if
no memory intercept / mallopt option is available.
This commit was SVN r18654.
about linkers, have all OPAL, ORTE, and OMPI components '''not'' link
against the OPAL, ORTE, or OMPI libraries.
See ttp://www.open-mpi.org/community/lists/users/2007/10/4220.php for
details (or https://svn.open-mpi.org/trac/ompi/wiki/Linkers for a
better-formatted version of the same info).
This commit was SVN r16968.
This merge adds Checkpoint/Restart support to Open MPI. The initial
frameworks and components support a LAM/MPI-like implementation.
This commit follows the risk assessment presented to the Open MPI core
development group on Feb. 22, 2007.
This commit closes trac:158
More details to follow.
This commit was SVN r14051.
The following SVN revisions from the original message are invalid or
inconsistent and therefore were not cross-referenced:
r13912
The following Trac tickets were found above:
Ticket 158 --> https://svn.open-mpi.org/trac/ompi/ticket/158
already set. This can annoy compilers that aren't GNUish
* __align is technically a reserved token and IBM XL appears to be doing
something with it that causes compile badness. So use a different
variable name.
refs trac:279
This commit was SVN r11264.
The following Trac tickets were found above:
Ticket 279 --> https://svn.open-mpi.org/trac/ompi/ticket/279
compiler, automatically disable the ptmalloc component. It seems that
optimization level -O2 or higher will cause the generated code to do
Bad Things (e.g., opalcc will segv). Upgrading to the Intel 9.1
compiler seems to fix the problem.
This closes ticket #227.
This commit was SVN r11076.
reentrant for free(), so we can't call free() from inside an sbrk() handler.
The solution is never call sbrk() with a negative number. The mmap() allocator
used for large allocations does not have this problem and continues to give
memory back to the OS as soon as possible.
This should go to both the v1.1 and v1.0 branches.
This commit was SVN r9943.
- move files out of toplevel include/ and etc/, moving it into the
sub-projects
- rather than including config headers with <project>/include,
have them as <project>
- require all headers to be included with a project prefix, with
the exception of the config headers ({opal,orte,ompi}_config.h
mpi.h, and mpif.h)
This commit was SVN r8985.
makes illegal free() calls behave in a much more rational way. You'll still
probably die, but your stack trace will not have 3 billion pages of recusion
inside the memory allocator.
* Fix illegal free in the opal_wrapper code. basename() returns a string in a
static buffer, so it shouldn't be free()ed. It also shouldn't be left around
so long, as another call to basename() may whack the returned buffer. So
leave the free and add a strdup() around the basename() call.
* Turn off some unneeded debugging in the opal_wrapper code that would list the
comamnd to be run, regardless of the -showme option.
This commit was SVN r8758.
page protection, which causes the pages to be droped, which causes problems
if we don't deregister the pages first. Since memory is cheap in this case
(it is still usable, should ptmalloc2 want it back, and is limited in size),
we just mprotect the pages instead. This solves the dropping pages problem,
and doesn't cause even more calls into the cache code.
Thanks to Gleb Natapov for both finding the problem and giving a fix.
This should go to the v1.0 branch
This commit was SVN r8732.
deallocation came from the allocator (malloc, fee, etc) or somewhere
else (the user calling mmap/munmap, etc). Going to be used by Galen
to determine if it is worth searching the allocations tree
Set flag if it is possible to intercept mmap (not always possible
due to a circular dependency between mmap, dlsym, and calloc)
This commit was SVN r8521.
may call calloc(large number), which causes ptmalloc2 to call mmap, which
causes us to try to dlsym for mmap, which leads to looping badness.
This commit was SVN r8461.
the ptmalloc2 memory hooks component triggers callbacks for memory
allocation / deallocation. If enabled (the default) it is only when
memory is actually obtained from or released to the OS (so little
malloc calls only trigger callbacks if sbrk is called). If disabled,
callbacks are triggered every time malloc/free/etc. is called
* It turns out that syscall and mmap aren't good friends due to the return
type of mmap and some old legacy issues with syscall functions that
take more than 5 parameters. For now, default to either loading
the symbol from glibc using dlsym or using the __m{un,}map functions.
Thanks to George for finding this.
* Fix some dumb typos in the mmap / munmap catching code
This commit was SVN r8410.
both mmap and munmap), adjusting the configure script so that the
component will only be activated on systems that use ptmalloc2 in the
first place -- ie, Linux
* Remove the malloc_hooks component - it became an unworkable solution
once threads and such were considered.
* Remove malloc_interpose component - it never worked quite right and
was not going to be able to intercept malloc, so it wasn't going to
be useful for OMPI's purposes.
* Update tests a little bit to match recent memory hooks api
issues - still needs a bit of work.
This commit was SVN r8381.
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.
AM_INIT_AUTOMAKE, instead of the deprecated version.
* Work around dumbness in modern AC_INIT that requires the version
number to be set at autoconf time (instead of at configure time, as
it was before). Set the version number, minus the subversion r number,
at autoconf time. Override the internal variables to include the r
number (if needed) at configure time. Basically, the right thing
should always happen. The only place it might not is the version
reported as part of configure --help will not have an r number.
* Since AM_INIT_AUTOMAKE taks a list of options, no need to specify
them in all the Makefile.am files.
* Addes support for subdir-objects, meaning that object files are put
in the directory containing source files, even if the Makefile.am is
in another directory. This should start making it feasible to
reduce the number of Makefile.am files we have in the tree, which
will greatly reduce the time to run autogen and configure.
This commit was SVN r7211.
on all glibc systems (tested with x86 and x86_64 with a couple of C++
compilers). While not as ideal as the malloc_hooks method, it does
have the advantage of working with threads.
* Modified malloc_hooks component to properly follow prefix rule. No
functionality changes
* Make the memory framework only chose one component, and modify all
components to set priority to 20, except malloc-interpose, which is
at 10. This means that on Linux, malloc_hooks will be used unless
threads are enabled, since I think malloc_hooks is a better design
choice when we can use it
This commit was SVN r6949.
* Add base to memory framework so that we can do something sane with
ompi_info
* Updated ompi_info to print components for memory framework and
show whether we have memory hooks active or not.
This commit was SVN r6861.
