For direct modex, all procs publish the selected pml module
and then at add_procs pml module for each proc is checked
against every other proc in the add_proc call.
For full modex, there is no change in functionality. Only Rank0
publishes its selected pml, all other procs in the add_proc call
check their selected pml against Rank0.
If pml's do not match, throw error and exit.
Signed-off-by: Dipti Kothari <dkothar@amazon.com>
Add a framework to support different types of threading models including
user space thread packages such as Qthreads and argobot:
https://github.com/pmodels/argobotshttps://github.com/Qthreads/qthreads
The default threading model is pthreads. Alternate thread models are
specificed at configure time using the --with-threads=X option.
The framework is static. The theading model to use is selected at
Open MPI configure/build time.
mca/threads: implement Argobots threading layer
config: fix thread configury
- Add double quotations
- Change Argobot to Argobots
config: implement Argobots check
If the poll time is too long, MPI hangs.
This quick fix just sets it to 0, but it is not good for the
Pthreads version. Need to find a good way to abstract it.
Note that even 1 (= 1 millisecond) causes disastrous performance
degradation.
rework threads MCA framework configury
It now works more like the ompi/mca/rte configury,
modulo some edge items that are special for threading package
linking, etc.
qthreads module
some argobots cleanup
Signed-off-by: Noah Evans <noah.evans@gmail.com>
Signed-off-by: Shintaro Iwasaki <siwasaki@anl.gov>
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
We currently save the hostname of a proc when we create the ompi_proc_t for it. This was originally done because the only method we had for discovering the host of a proc was to include that info in the modex, and we had to therefore store it somewhere proc-local. Obviously, this ccarried a memory penalty for storing all those strings, and so we added a "cutoff" parameter so that we wouldn't collect hostnames above a certain number of procs.
Unfortunately, this still results in an 8-byte/proc memory cost as we have a char* pointer in the opal_proc_t that is contained in the ompi_proc_t so that we can store the hostname of the other procs if we fall below the cutoff. At scale, this can consume a fair amount of memory.
With the switch to relying on PMIx, there is no longer a need to cache the proc hostnames. Using the "optional" feature of PMIx_Get, we restrict the retrieval to be purely proc-local - i.e., we retrieve the info either via shared memory or from within the proc-internal hash storage (depending upon the active PMIx components). Thus, the retrieval of a hostname is purely a local operation involving no communication.
All RM's are required to provide a complete hostname map of all procs at startup. Thus, we have full access to all hostnames without including them in a modex or having to cache them on each proc. This allows us to remove the char* pointer from the opal_proc_t, saving us 8-bytes/proc.
Unfortunately, PMIx_Get does not currently support the return of a static pointer to memory. Thus, even though PMIx has the hostname in its memory, it can only return a malloc'd version of it. I have therefore ensured that the return from opal_get_proc_hostname is consistently malloc'd and free'd wherever used. This shouldn't be a burden as the hostname is only used in one of two circumstances:
(a) in an error message
(b) in a verbose output for debugging purposes
Thus, there should be no performance penalty associated with the malloc/free requirement. PMIx will eventually be returning static pointers, and so we can eventually simplify this method and return a "const char*" - but as noted, this really isn't an issue even today.
Signed-off-by: Ralph Castain <rhc@pmix.org>
Temporary solution for the PML inconsistency issue discussed in #7475.
This patch address 2 things: first it make the PMIx key optional so that
if we are not in a full modex mode we don't do a direct modex, and
second it get the PML info from the vpid 0 instead of from the local
rank.
Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
Will be replaced by PRRTE. Ensure that OMPI and OPAL layers build
without reference to ORTE. Setup opal/pmix framework to be static.
Remove support for all PMI-1 and PMI-2 libraries. Add support for
"external" pmix component as well as internal v4 one.
remove orte: misc fixes
- UCX fixes
- VPATH issue
- oshmem fixes
- remove useless definition
- Add PRRTE submodule
- Get autogen.pl to traverse PRRTE submodule
- Remove stale orcm reference
- Configure embedded PRRTE
- Correctly pass the prefix to PRRTE
- Correctly set the OMPI_WANT_PRRTE am_conditional
- Move prrte configuration to the end of OMPI's configure.ac
- Make mpirun a symlink to prun, when available
- Fix makedist with --no-orte/--no-prrte option
- Add a `--no-prrte` option which is the same as the legacy
`--no-orte` option.
- Remove embedded PMIx tarball. Replace it with new submodule
pointing to OpenPMIx master repo's master branch
- Some cleanup in PRRTE integration and add config summary entry
- Correctly set the hostname
- Fix locality
- Fix singleton operations
- Fix support for "tune" and "am" options
Signed-off-by: Ralph Castain <rhc@pmix.org>
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
Signed-off-by: Joshua Hursey <jhursey@us.ibm.com>
This is closely related to Platform-MPI's old -prot feature.
The long-format of the tables it prints could look like this:
> Host 0 [myhost001] ranks 0 - 1
> Host 1 [myhost002] ranks 2 - 3
> Host 2 [myhost003] ranks 4
> Host 3 [myhost004] ranks 5
> Host 4 [myhost005] ranks 6
> Host 5 [myhost006] ranks 7
> Host 6 [myhost007] ranks 8
> Host 7 [myhost008] ranks 9
> Host 8 [myhost009] ranks 10
>
> host | 0 1 2 3 4 5 6 7 8
> ======|==============================================
> 0 : sm tcp tcp tcp tcp tcp tcp tcp tcp
> 1 : tcp sm tcp tcp tcp tcp tcp tcp tcp
> 2 : tcp tcp self tcp tcp tcp tcp tcp tcp
> 3 : tcp tcp tcp self tcp tcp tcp tcp tcp
> 4 : tcp tcp tcp tcp self tcp tcp tcp tcp
> 5 : tcp tcp tcp tcp tcp self tcp tcp tcp
> 6 : tcp tcp tcp tcp tcp tcp self tcp tcp
> 7 : tcp tcp tcp tcp tcp tcp tcp self tcp
> 8 : tcp tcp tcp tcp tcp tcp tcp tcp self
>
> Connection summary:
> on-host: all connections are sm or self
> off-host: all connections are tcp
In this example hosts 0 and 1 had multiple ranks so "sm" was more
meaningful than "self" to identify how the ranks on the host are
talking to each other. While host 2..8 were one rank per host so
"self" was more meaningful as their btl.
Above a certain number of hosts (12 by default) the above table gets too big
so we shrink to a more abbreviated looking table that has the same data:
> host | 0 1 2 3 4 8
> ======|====================
> 0 : A C C C C C C C C
> 1 : C A C C C C C C C
> 2 : C C B C C C C C C
> 3 : C C C B C C C C C
> 4 : C C C C B C C C C
> 5 : C C C C C B C C C
> 6 : C C C C C C B C C
> 7 : C C C C C C C B C
> 8 : C C C C C C C C B
> key: A == sm
> key: B == self
> key: C == tcp
Then above 36 hosts we stop printing the 2d table entirely and just print the
summary:
> Connection summary:
> on-host: all connections are sm or self
> off-host: all connections are tcp
The options to control it are
-mca comm_method 1 : print the above table at the end of MPI_Init
-mca comm_method 2 : print the above table at the beginning of MPI_Finalize
-mca comm_method_max <n> : number of hosts <n> for which to print a full size 2d
-mca comm_method_brief 1 : only print summary output, no 2d table
-mca comm_method_fakefile <filename> : for debugging only
* printing at init vs finalize:
The most important difference between these two is that when printing the table
during MPI_Init(), we send extra messages to make sure all hosts are connected to
each other. So the table ends up working against the idea of on-demand connections
(although it's only forcing the n^2 connections in the number of hosts, not the
total ranks). If printing at MPI_Finalize() we don't create any connections that
aren't already connected, so the table is more likely to have "n/a" entries if
some hosts never connected to each other.
* how many hosts <n> for which to print a full size 2d table
The option -mca comm_method_max <n> can be used to specify a number of hosts <n>
(default 12) that controls at what host-count the unabbreviated / abbreviated
2d tables get printed:
1 - n : full size 2d table
n+1 - 3n : shortened 2d table
3n+1 - inf : summary only, no 2d table
* brief
The option -mca comm_method_brief 1 can be used to skip the printing of the 2d
table and only show the short summary
* fakefile
This is a debugging option that allows easeir testing of all the printout
routines by letting all the detected communication methods between the hosts
be overridden by fake data from a file.
The source of the information used in the table is the .mca_component_name
In the case of BTLs, the module always had a .btl_component linking back to the
component. The vars mca_pml_base_selected_component and ompi_mtl_base_selected_component
offer similar functionality for pml/mtl.
So with the ability to identify the component, we can then access
the component name with code like this
mca_pml_base_selected_component.pmlm_version.mca_component_name
See the three lookup_{pml,mtl,btl}_name() functions in hook_comm_method_fns.c,
and their use in comm_method() to parse the strings and produce an integer
to represent the connection type being used.
Signed-off-by: Mark Allen <markalle@us.ibm.com>
With this patch the best PML is selected earlier, before finalizing
the others PML. This provides a simpler mechanism to intercept and
highjack the PML (as done in the monitoring PML)
Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
This commit updates the entire codebase to use specific opal types for
all atomic variables. This is a change from the prior atomic support
which required the use of the volatile keyword. This is the first step
towards implementing support for C11 atomics as that interface
requires the use of types declared with the _Atomic keyword.
Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
This commit renames the arithmetic atomic operations in opal to
indicate that they return the new value not the old value. This naming
differentiates these routines from new functions that return the old
value.
Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
On start we were not correctly resetting all request fields. This was
leading to a double-completion on persistent receives. This commit
updates the base start code to reset the receive req_bytes_packed and
the send request convertor.
Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
* Remodel the request.
Added the wait sync primitive and integrate it into the PML and MTL
infrastructure. The multi-threaded requests are now significantly
less heavy and less noisy (only the threads associated with completed
requests are signaled).
* Fix the condition to release the request.
This commit rewrites both the mpool and rcache frameworks. Summary of
changes:
- Before this change a significant portion of the rcache
functionality lived in mpool components. This meant that it was
impossible to add a new memory pool to use with rdma networks
(ugni, openib, etc) without duplicating the functionality of an
existing mpool component. All the registration functionality has
been removed from the mpool and placed in the rcache framework.
- All registration cache mpools components (udreg, grdma, gpusm,
rgpusm) have been changed to rcache components. rcaches are
allocated and released in the same way mpool components were.
- It is now valid to pass NULL as the resources argument when
creating an rcache. At this time the gpusm and rgpusm components
support this. All other rcache components require non-NULL
resources.
- A new mpool component has been added: hugepage. This component
supports huge page allocations on linux.
- Memory pools are now allocated using "hints". Each mpool component
is queried with the hints and returns a priority. The current hints
supported are NULL (uses posix_memalign/malloc), page_size=x (huge
page mpool), and mpool=x.
- The sm mpool has been moved to common/sm. This reflects that the sm
mpool is specialized and not meant for any general
allocations. This mpool may be moved back into the mpool framework
if there is any objection.
- The opal_free_list_init arguments have been updated. The unused0
argument is not used to pass in the registration cache module. The
mpool registration flags are now rcache registration flags.
- All components have been updated to make use of the new framework
interfaces.
As this commit makes significant changes to both the mpool and rcache
frameworks both versions have been bumped to 3.0.0.
Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
Bring Slurm PMI-1 component online
Bring the s2 component online
Little cleanup - let the various PMIx modules set the process name during init, and then just raise it up to the ORTE level. Required as the different PMI environments all pass the jobid in different ways.
Bring the OMPI pubsub/pmi component online
Get comm_spawn working again
Ensure we always provide a cpuset, even if it is NULL
pmix/cray: adjust cray pmix component for pmix
Make changes so cray pmix can work within the integrated
ompi/pmix framework.
Bring singletons back online. Implement the comm_spawn operation using pmix - not tested yet
Cleanup comm_spawn - procs now starting, error in connect_accept
Complete integration
This commit does two things. It removes checks for C99 required
headers (stdlib.h, string.h, signal.h, etc). Additionally it removes
definitions for required C99 types (intptr_t, int64_t, int32_t, etc).
Signed-off-by: Nathan Hjelm <hjelmn@me.com>
This commit adds support for project_framework_component_* parameter
matching. This is the first step in allowing the same framework name
in multiple projects. This change also bumps the MCA component version
to 2.1.0.
All master frameworks have been updated to use the new component
versioning macro. An mca.h has been added to each project to add a
project specific versioning macro of the form
PROJECT_MCA_VERSION_2_1_0.
Signed-off-by: Nathan Hjelm <hjelmn@me.com>
Use of the old ompi_free_list_t and ompi_free_list_item_t is
deprecated. These classes will be removed in a future commit.
This commit updates the entire code base to use opal_free_list_t and
opal_free_list_item_t.
Notes:
OMPI_FREE_LIST_*_MT -> opal_free_list_* (uses opal_using_threads ())
Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
This commit adds an owner file in each of the component directories
for each framework. This allows for a simple script to parse
the contents of the files and generate, among other things, tables
to be used on the project's wiki page. Currently there are two
"fields" in the file, an owner and a status. A tool to parse
the files and generate tables for the wiki page will be added
in a subsequent commit.
WHAT: Merge the PMIx branch into the devel repo, creating a new
OPAL “lmix” framework to abstract PMI support for all RTEs.
Replace the ORTE daemon-level collectives with a new PMIx
server and update the ORTE grpcomm framework to support
server-to-server collectives
WHY: We’ve had problems dealing with variations in PMI implementations,
and need to extend the existing PMI definitions to meet exascale
requirements.
WHEN: Mon, Aug 25
WHERE: https://github.com/rhc54/ompi-svn-mirror.git
Several community members have been working on a refactoring of the current PMI support within OMPI. Although the APIs are common, Slurm and Cray implement a different range of capabilities, and package them differently. For example, Cray provides an integrated PMI-1/2 library, while Slurm separates the two and requires the user to specify the one to be used at runtime. In addition, several bugs in the Slurm implementations have caused problems requiring extra coding.
All this has led to a slew of #if’s in the PMI code and bugs when the corner-case logic for one implementation accidentally traps the other. Extending this support to other implementations would have increased this complexity to an unacceptable level.
Accordingly, we have:
* created a new OPAL “pmix” framework to abstract the PMI support, with separate components for Cray, Slurm PMI-1, and Slurm PMI-2 implementations.
* Replaced the current ORTE grpcomm daemon-based collective operation with an integrated PMIx server, and updated the grpcomm APIs to provide more flexible, multi-algorithm support for collective operations. At this time, only the xcast and allgather operations are supported.
* Replaced the current global collective id with a signature based on the names of the participating procs. The allows an unlimited number of collectives to be executed by any group of processes, subject to the requirement that only one collective can be active at a time for a unique combination of procs. Note that a proc can be involved in any number of simultaneous collectives - it is the specific combination of procs that is subject to the constraint
* removed the prior OMPI/OPAL modex code
* added new macros for executing modex send/recv to simplify use of the new APIs. The send macros allow the caller to specify whether or not the BTL supports async modex operations - if so, then the non-blocking “fence” operation is used, if the active PMIx component supports it. Otherwise, the default is a full blocking modex exchange as we currently perform.
* retained the current flag that directs us to use a blocking fence operation, but only to retrieve data upon demand
This commit was SVN r32570.
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.
When compiling --with-ft there are a few compiler warnings about
unused variables. This patch fixes those compiler warnings.
This commit was SVN r30927.
* add a new MCA param orte_hostname_cutoff to specify the number of nodes at which we stop including hostnames. This defaults to INT_MAX => always include hostnames. If a value is given, then we will include hostnames for any allocation smaller than the given limit.
* remove ompi_proc_get_hostname. Replace all occurrences with a direct link to ompi_proc_t's proc_hostname, protected by appropriate "if NULL"
* modify the OMPI-ORTE integration component so that any call to modex_recv automatically loads the ompi_proc_t->proc_hostname field as well as returning the requested info. Thus, any process whose modex info you retrieve will automatically receive the hostname. Note that on-demand retrieval is still enabled - i.e., if we are running under direct launch with PMI, the hostname will be fetched upon first call to modex_recv, and then the ompi_proc_t->proc_hostname field will be loaded
* removed a stale MCA param "mpi_keep_peer_hostnames" that was no longer used anywhere in the code base
* added an envar lookup in ess/pmi for the number of nodes in the allocation. Sadly, PMI itself doesn't provide that info, so we have to get it a different way. Currently, we support PBS-based systems and SLURM - for any other, rank0 will emit a warning and we assume max number of daemons so we will always retain hostnames
This commit was SVN r29052.
This creates a really bad scaling behavior. Users have found a nearly 20% launch time differential between mpirun and PMI, with PMI being the slower method. Some of the problem is attributable to poor exchange algorithms in RM's like Slurm and Alps, but we make things worse by calling "get" so many times.
Nathan (with a tad advice from me) has attempted to alleviate this problem by reducing the number of "get" calls. This required the following changes:
* upon first request for data, have the OPAL db pmi component fetch and decode *all* the info from a given remote proc. It turned out we weren't caching the info, so we would continually request it and only decode the piece we needed for the immediate request. We now decode all the info and push it into the db hash component for local storage - and then all subsequent retrievals are fulfilled locally
* reduced the amount of data by eliminating the exchange of the OMPI_ARCH value if heterogeneity is not enabled. This was used solely as a check so we would error out if the system wasn't actually homogeneous, which was fine when we thought there was no cost in doing the check. Unfortunately, at large scale and with direct launch, there is a non-zero cost of making this test. We are open to finding a compromise (perhaps turning the test off if requested?), if people feel strongly about performing the test
* reduced the amount of RTE data being automatically fetched, and fetched the rest only upon request. In particular, we no longer immediately fetch the hostname (which is only used for error reporting), but instead get it when needed. Likewise for the RML uri as that info is only required for some (not all) environments. In addition, we no longer fetch the locality unless required, relying instead on the PMI clique info to tell us who is on our local node (if additional info is required, the fetch is performed when a modex_recv is issued).
Again, all this only impacts direct launch - all the info is provided when launched via mpirun as there is no added cost to getting it
Barring objections, we may move this (plus any required other pieces) to the 1.7 branch once it soaks for an appropriate time.
This commit was SVN r29040.
Notes:
- This commit also eliminates the need for an available components list in use
in several frameworks. None of the code in question was making use of the
priority field of the priority component list item so these extra lists were
removed.
- Cleaned up selection code in several frameworks to sort lists using opal_list_sort.
- Cleans up the ompi/orte-info functions. Expose the functions that construct the
list of params so they can be used elsewhere.
patches for mtl/portals4 from brian
missed a few output variables in openib
This commit was SVN r28241.
Features:
- Support for an override parameter file (openmpi-mca-param-override.conf).
Variable values in this file can not be overridden by any file or environment
value.
- Support for boolean, unsigned, and unsigned long long variables.
- Support for true/false values.
- Support for enumerations on integer variables.
- Support for MPIT scope, verbosity, and binding.
- Support for command line source.
- Support for setting variable source via the environment using
OMPI_MCA_SOURCE_<var name>=source (either command or file:filename)
- Cleaner API.
- Support for variable groups (equivalent to MPIT categories).
Notes:
- Variables must be created with a backing store (char **, int *, or bool *)
that must live at least as long as the variable.
- Creating a variable with the MCA_BASE_VAR_FLAG_SETTABLE enables the use of
mca_base_var_set_value() to change the value.
- String values are duplicated when the variable is registered. It is up to
the caller to free the original value if necessary. The new value will be
freed by the mca_base_var system and must not be freed by the user.
- Variables with constant scope may not be settable.
- Variable groups (and all associated variables) are deregistered when the
component is closed or the component repository item is freed. This
prevents a segmentation fault from accessing a variable after its component
is unloaded.
- After some discussion we decided we should remove the automatic registration
of component priority variables. Few component actually made use of this
feature.
- The enumerator interface was updated to be general enough to handle
future uses of the interface.
- The code to generate ompi_info output has been moved into the MCA variable
system. See mca_base_var_dump().
opal: update core and components to mca_base_var system
orte: update core and components to mca_base_var system
ompi: update core and components to mca_base_var system
This commit also modifies the rmaps framework. The following variables were
moved from ppr and lama: rmaps_base_pernode, rmaps_base_n_pernode,
rmaps_base_n_persocket. Both lama and ppr create synonyms for these variables.
This commit was SVN r28236.
