Don't use "ompi" names with top-level constructs, because the OMPI
project may not be built. Instead, just use "opal" names -- they'll
always be there.
This commit was SVN r32446.
WHAT: Open our low-level communication infrastructure by moving all necessary components (btl/rcache/allocator/mpool) down in OPAL
All the components required for inter-process communications are currently deeply integrated in the OMPI layer. Several groups/institutions have express interest in having a more generic communication infrastructure, without all the OMPI layer dependencies. This communication layer should be made available at a different software level, available to all layers in the Open MPI software stack. As an example, our ORTE layer could replace the current OOB and instead use the BTL directly, gaining access to more reactive network interfaces than TCP. Similarly, external software libraries could take advantage of our highly optimized AM (active message) communication layer for their own purpose. UTK with support from Sandia, developped a version of Open MPI where the entire communication infrastucture has been moved down to OPAL (btl/rcache/allocator/mpool). Most of the moved components have been updated to match the new schema, with few exceptions (mainly BTLs where I have no way of compiling/testing them). Thus, the completion of this RFC is tied to being able to completing this move for all BTLs. For this we need help from the rest of the Open MPI community, especially those supporting some of the BTLs. A non-exhaustive list of BTLs that qualify here is: mx, portals4, scif, udapl, ugni, usnic.
This commit was SVN r32317.
new automake requires subdirs-object directive, to resolve this:
09:43:37 automake: warning: possible forward-incompatibility.
09:43:37 automake: At least a source file is in a subdirectory, but the 'subdir-objects'
09:43:37 automake: automake option hasn't been enabled. For now, the corresponding output
09:43:37 automake: object file(s) will be placed in the top-level directory. However,
09:43:37 automake: this behaviour will change in future Automake versions: they will
09:43:37 automake: unconditionally cause object files to be placed in the same subdirectory
09:43:37 automake: of the corresponding sources.
09:43:37 automake: You are advised to start using 'subdir-objects' option throughout your
09:43:37 automake: project, to avoid future incompatibilities.
09:43:37 tools/otfmerge/Makefile.common:13: warning: source file '$(OTFMERGESRCDIR)/otfmerge.c' is in a subdirectory,
09:43:37 tools/otfmerge/Makefile.common:13: but option 'subdir-objects' is disabled
cmr=v1.8.2:reviewer=ompi-rm1.8
This commit was SVN r32225.
consistency across different distros: src.rpm create on one distro can be unpacked on
another when same packing algorithm used on both.
reviewed by Jeff
cmr=v1.8.2:reviewer=ompi-rm1.8
This commit was SVN r32205.
Update the Open MPI description and fix lots of grammatical errors in
the OpenSHMEM description.
cmr=v1.8.2:reviewer=ompi-rm1.8
This commit was SVN r31709.
* Ensure to cd to the basename of the URL (e.g., "v1.8" from
"branches/v1.8")
* Remove the erroneous -m from the svn commit line that was
preventing the updated .gitignore file from getting committed
This commit was SVN r31548.
Add some getopt options for both build-gitignore.pl and
build-hgignore.pl:
-v: be a little verbose
-h: show a quick help message
-o: specify the output file to write
build-hgignore.pl defaults to writing .hgignore, and
build-gitignore.pl defaults to writing .gitignore. But you can use -o
to specify writing a different filename.
This commit was SVN r31397.
Sort the output from reading directory contents so that the generated
.gitignore and .hgignore files will be much more consistent (and
friendly to change from subsequent runs, especially when they are
committed to SVN/git/hg repos).
This commit was SVN r31394.
(i.e., the script that fires via cron every 15 mins to "git pull" from
the github ompi-www repo).
This commit moves the mercurial mirroring script to the
"infrastructure" directory.
This commit was SVN r31360.
really "nightly" scripting.
Also, the crontab files were already moved to the infrastructure
directory; forgot to commit their removals from the "nightly"
directory.
This commit was SVN r31359.
completion from _arguments instead of treating it as a state
cmr=v1.8.1:ticket=trac:4500
This commit was SVN r31350.
The following Trac tickets were found above:
Ticket 4500 --> https://svn.open-mpi.org/trac/ompi/ticket/4500
completion
_normal forces completion to be done on the subcommand. This gives us two
things:
- Completion of an executable for the first command argument
- Completion of options for that executable if any exist. If not it
will use normal completion rules (_files).
cmr=v1.8.1:reviewer=jsquyres
This commit was SVN r31346.
This script can be run from any sub-tree in the OMPI code base. It
does several things:
* Find all help-*txt files and index all the topics found in the
sub-tree.
* Find all C/C++ source files (.c, .cc, .h) in the sub-tree, and look
for various flavors of the opal_show_help() function (e.g.,
including orte_show_help()) and search for hard-coded filenames and
topics.
* Also look for special tokens (in comments) in the source code for
help topics that are not necessarily hard-coded (e.g., topics that
are snprintf'ed).
* For each filename/topic found, output a warning if a) the file does
not exist, or b) that the topic does not exist in that file.
* Output a warning for any topic that is not referenced in the source
code (i.e., orphaned/now-unused help messages).
* Output a warning for any help file that appears to be empty.
* Output a warning for any help file that appears to be unused (i.e.,
no topics in the help file are referenced in code).
This commit was SVN r31311.
Also, since I put some of the macros for these silent/verbose rules up
in the top-level Makefile.man-page-rules file, I renamed it to
Makefile.ompi-rules.
I've had this sitting around for a while; now seems like as good a
time as any to commit it.
This commit was SVN r31271.
The gatekeeper script was not correctly respecting the locale specified
in the user's environment. So basically this scenario could (and did)
easily happen:
1. A committer writes a valid message in UTF-8 and runs `svn commit` with
a correct locale setting of `LANG=en_US.UTF-8`.
2. SVN transcodes that to UTF-8 for internal storage (a no-op in this
case).
3. The gatekeeper, also with `LANG=en_US.UTF-8` set, runs
`gkcommit.pl ...`. This breaks down into the following steps:
A. run `svn log --xml ...`, which SVN correctly transcodes from UTF-8
into the current locale, which happens to also be UTF-8
B. Perl reads this in and assumes this is a sequence of raw 8-bit
bytes in a "native" latin1-type encoding.
C. Perl's XML::Parser module spots the XML declaration stating the
content is UTF-8 encoded: `<?xml version="1.0" encoding="UTF-8"?>`.
Perl internally stores the parsed strings as proper Unicode
strings (UTF-8 encoded internally, but that's irrelevant here).
D. Perl writes out the commit message file in the _latin1_ encoding,
transcoding characters from internal UTF-8. This causes
characters like "ä" (Unicode code point: 0xe4, UTF-8 encoding:
0xc3 0xa4) to be encoded as a single byte: 0xe4.
This fix changes the behavior at steps 3A and 3D to transparently treat
the incoming/outgoing data as UTF-8 (assuming a UTF-8 locale is set in
the user's environment).
There can still be problems if either the committer or the gatekeeper
have locale settings that do not agree with the encoding that their
editor is producing, but such is i18n :(
Helpful references for anyone debugging this sort of issue in the
future:
* http://perldoc.perl.org/perllocale.html#Unicode-and-UTF-8
* http://perldoc.perl.org/perluniintro.html#Unicode-I%2fO
Refs trac:4217
Reviewed-by: Jeff Squyres <jsquyres@cisco.com>
cmr=v1.7.5:reviewer=ompi-rm1.7
This commit was SVN r30709.
The following Trac tickets were found above:
Ticket 4217 --> https://svn.open-mpi.org/trac/ompi/ticket/4217
pkg{data,lib,includedir}, use our own ompi{data,lib,includedir}, which is
always set to {datadir,libdir,includedir}/openmpi. This will keep us from
having help files in prefix/share/open-rte when building without Open MPI,
but in prefix/share/openmpi when building with Open MPI.
This commit was SVN r30140.
To support the new mpool two changes were made to the mpool infrastructure:
1) Added an mpool flag to indicate that an mpool does not need the memory
hooks to use the leave pinned protocols. This flag is checked in the
mpool lookup.
2) Add a mpool context to the base registration. This new member is used
by the udreg mpool to store the udreg context associated with the
particular registration. The new member will not break the ABI
compatibility as the new member is only currently used by the udreg
mpool.
Dynamics support for Cray systems makes use of the global rank provided by
orte to give the ugni library a unique rank for each process. Dynamics
support is not available under direct-launch (srun.)
cmr=v1.7.4
This commit was SVN r29719.
unfortunately the debian packaging files should reside in the root folder
and cannot be placed under contrib/dist/... tree.
developed by Aleksey, reviewed by miked
cmr=v1.7.4:reviewer=ompi-gk1.7
This commit was SVN r29615.
Features of v 1.0:
- Completion of all switches.
- Completion of mca variable names.
- Completion of mca variable values for enumerated variables and component
selection variables.
- Completion of --bind-to and --map-by.
This commit was SVN r29513.
*** THIS RFC INCLUDES A MINOR CHANGE TO THE MPI-RTE INTERFACE ***
Note: during the course of this work, it was necessary to completely separate the MPI and RTE progress engines. There were multiple places in the MPI layer where ORTE_WAIT_FOR_COMPLETION was being used. A new OMPI_WAIT_FOR_COMPLETION macro was created (defined in ompi/mca/rte/rte.h) that simply cycles across opal_progress until the provided flag becomes false. Places where the MPI layer blocked waiting for RTE to complete an event have been modified to use this macro.
***************************************************************************************
I am reissuing this RFC because of the time that has passed since its original release. Since its initial release and review, I have debugged it further to ensure it fully supports tests like loop_spawn. It therefore seems ready for merge back to the trunk. Given its prior review, I have set the timeout for one week.
The code is in https://bitbucket.org/rhc/ompi-oob2
WHAT: Rewrite of ORTE OOB
WHY: Support asynchronous progress and a host of other features
WHEN: Wed, August 21
SYNOPSIS:
The current OOB has served us well, but a number of limitations have been identified over the years. Specifically:
* it is only progressed when called via opal_progress, which can lead to hangs or recursive calls into libevent (which is not supported by that code)
* we've had issues when multiple NICs are available as the code doesn't "shift" messages between transports - thus, all nodes had to be available via the same TCP interface.
* the OOB "unloads" incoming opal_buffer_t objects during the transmission, thus preventing use of OBJ_RETAIN in the code when repeatedly sending the same message to multiple recipients
* there is no failover mechanism across NICs - if the selected NIC (or its attached switch) fails, we are forced to abort
* only one transport (i.e., component) can be "active"
The revised OOB resolves these problems:
* async progress is used for all application processes, with the progress thread blocking in the event library
* each available TCP NIC is supported by its own TCP module. The ability to asynchronously progress each module independently is provided, but not enabled by default (a runtime MCA parameter turns it "on")
* multi-address TCP NICs (e.g., a NIC with both an IPv4 and IPv6 address, or with virtual interfaces) are supported - reachability is determined by comparing the contact info for a peer against all addresses within the range covered by the address/mask pairs for the NIC.
* a message that arrives on one TCP NIC is automatically shifted to whatever NIC that is connected to the next "hop" if that peer cannot be reached by the incoming NIC. If no TCP module will reach the peer, then the OOB attempts to send the message via all other available components - if none can reach the peer, then an "error" is reported back to the RML, which then calls the errmgr for instructions.
* opal_buffer_t now conforms to standard object rules re OBJ_RETAIN as we no longer "unload" the incoming object
* NIC failure is reported to the TCP component, which then tries to resend the message across any other available TCP NIC. If that doesn't work, then the message is given back to the OOB base to try using other components. If all that fails, then the error is reported to the RML, which reports to the errmgr for instructions
* obviously from the above, multiple OOB components (e.g., TCP and UD) can be active in parallel
* the matching code has been moved to the RML (and out of the OOB/TCP component) so it is independent of transport
* routing is done by the individual OOB modules (as opposed to the RML). Thus, both routed and non-routed transports can simultaneously be active
* all blocking send/recv APIs have been removed. Everything operates asynchronously.
KNOWN LIMITATIONS:
* although provision is made for component failover as described above, the code for doing so has not been fully implemented yet. At the moment, if all connections for a given peer fail, the errmgr is notified of a "lost connection", which by default results in termination of the job if it was a lifeline
* the IPv6 code is present and compiles, but is not complete. Since the current IPv6 support in the OOB doesn't work anyway, I don't consider this a blocker
* routing is performed at the individual module level, yet the active routed component is selected on a global basis. We probably should update that to reflect that different transports may need/choose to route in different ways
* obviously, not every error path has been tested nor necessarily covered
* determining abnormal termination is more challenging than in the old code as we now potentially have multiple ways of connecting to a process. Ideally, we would declare "connection failed" when *all* transports can no longer reach the process, but that requires some additional (possibly complex) code. For now, the code replicates the old behavior only somewhat modified - i.e., if a module sees its connection fail, it checks to see if it is a lifeline. If so, it notifies the errmgr that the lifeline is lost - otherwise, it notifies the errmgr that a non-lifeline connection was lost.
* reachability is determined solely on the basis of a shared subnet address/mask - more sophisticated algorithms (e.g., the one used in the tcp btl) are required to handle routing via gateways
* the RML needs to assign sequence numbers to each message on a per-peer basis. The receiving RML will then deliver messages in order, thus preventing out-of-order messaging in the case where messages travel across different transports or a message needs to be redirected/resent due to failure of a NIC
This commit was SVN r29058.
