1
1
Граф коммитов

32 Коммитов

Автор SHA1 Сообщение Дата
George Bosilca
24879f9def Code cleanup while chasing valgrind complaints.
This commit was SVN r30048.
2013-12-21 23:28:14 +00:00
Ralph Castain
9902748108 ***** THIS INCLUDES A SMALL CHANGE IN THE MPI-RTE INTERFACE *****
Fix two problems that surfaced when using direct launch under SLURM:

1. locally store our own data because some BTLs want to retrieve 
   it during add_procs rather than use what they have internally

2. cleanup MPI_Abort so it correctly passes the error status all
   the way down to the actual exit. When someone implemented the
   "abort_peers" API, they left out the error status. So we lost
   it at that point and *always* exited with a status of 1. This 
   forces a change to the API to include the status.

cmr:v1.7.3:reviewer=jsquyres:subject=Fix MPI_Abort and modex_recv for direct launch

This commit was SVN r29405.
2013-10-08 18:37:59 +00:00
Ralph Castain
2121e9c01b Fix an issue regarding use of PMI when running processes and tools that don't need or want to use it. We build PMI support based on configuration settings and library availability.
However, tools such as mpirun don't need it, and definitely shouldn't be using it. Ditto for procs launched by mpirun.

We used to have a way of dealing with this - we had the PMI component check to see if the process was the HNP or was launched by an HNP. Sadly, moving the OPAL db framework removed
 that ability as OPAL has no notion of HNPs or proc type.

So add a boolean flag to the db_base_select API that allows us to restrict selection to "local" components. This gives the PMI component the ability to reject itself as required. W
e then need to pass that param into the ess_base_std_app call so it can pass it all down.

This commit was SVN r29341.
2013-10-02 19:03:46 +00:00
Ralph Castain
d565a76814 Do some cleanup of the way we handle modex data. Identify data that needs to be shared with peers in my job vs data that needs to be shared with non-peers - no point in sharing extra data. When we share data with some process(es) from another job, we cannot know in advance what info they have or lack, so we have to share everything just in case. This limits the optimization we can do for things like comm_spawn.
Create a new required key in the OMPI layer for retrieving a "node id" from the database. ALL RTE'S MUST DEFINE THIS KEY. This allows us to compute locality in the MPI layer, which is necessary when we do things like intercomm_create.

cmr:v1.7.4:reviewer=rhc:subject=Cleanup handling of modex data

This commit was SVN r29274.
2013-09-27 00:37:49 +00:00
Ralph Castain
9aeba777fa Ensure we don't enter into an infinite loop looking for the PML modex key if it isn't present. The PMI implementation will load ALL modex keys when the first key is queried, so the hash db component can safely return "not found" if a subsequent key isn't present. The PML modex_recv needs to assume everything is okay if the modex recv fails to return a value.
cmr:v1.7.3:reviewer=jladd:subject=Prevent infinite loop when PML modex not found

This commit was SVN r29243.
2013-09-25 16:04:00 +00:00
Ralph Castain
63da76ad5f Silence warnings about pointer casting
This commit was SVN r29226.
2013-09-22 19:21:29 +00:00
Nathan Hjelm
bc31773523 Fix bug in db/pmi when a stored byte object has a NULL pointer.
cmr=v1.7.3:reviewer=samuel

This commit was SVN r29215.
2013-09-20 15:38:36 +00:00
Ralph Castain
2245ac0e7e Don't error log the return from setup_pmi as it can indicate that the process wasn't launched via srun or its equivalent.
cmr:v1.7.3:reviewer=jladd

This commit was SVN r29180.
2013-09-17 02:26:46 +00:00
Alex Margolin
50a3c01a0f fixed build without thread support
This commit was SVN r29145.
2013-09-06 19:03:19 +00:00
Ralph Castain
3516348aad We don't need to report errors in pmi_setup as it is possible that PMI is available, but that we weren't launched under it (e.g., we launched via mpirun).
cmr:v1.7.3:reviewer=hjelmn:subject="Silence unnecessary PMI error msgs"

This commit was SVN r29086.
2013-08-29 16:35:20 +00:00
Ralph Castain
a200e4f865 As per the RFC, bring in the ORTE async progress code and the rewrite of OOB:
*** 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.
2013-08-22 16:37:40 +00:00
Ralph Castain
611d7f9f6b When we direct launch an application, we rely on PMI for wireup support. In doing so, we lose the de facto data compression we get from the ORTE modex since we no longer get all the wireup info from every proc in a single blob. Instead, we have to iterate over all the procs, calling PMI_KVS_get for every value we require.
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.
2013-08-17 00:49:18 +00:00
Nathan Hjelm
be1bd4661c db/pmi: speed up modex by caching pmi data internally
This commit was SVN r29001.
2013-08-05 22:31:50 +00:00
Nathan Hjelm
88cadc552d Make opal/db/pmi use as few PMI keys as possible.
This commit reintroduces key compression into the pmi db. This feature
compresses the keys stored into the component into a small number of
PMI keys by serializing the data and base64 encoding the result. This
will avoid issues with Cray PMI which restricts us to ~ 3 PMI keys per
rank.

This commit was SVN r28993.
2013-08-03 01:06:59 +00:00
Ralph Castain
3c8aa7c296 Don't just hardcode the max length of the PMI name as it could be wrong. PMI2 installations seem to be retaining at least some of the PMI functions, so use the one to get the max name length.
This commit was SVN r28962.
2013-07-30 14:13:15 +00:00
Ralph Castain
41f97931e9 Need to include module-level CPPFLAGS so it can build
This commit was SVN r28947.
2013-07-24 23:07:43 +00:00
Ralph Castain
6c1a140e99 Per request from Nathan, add a "commit" API to the opal db framework. This allows him to aggregate keys to work around the Cray's severe PMI limitations
This commit was SVN r28917.
2013-07-22 22:57:16 +00:00
Ralph Castain
028f5ee7a6 Cleanup some bitrot from moving the db framework to opal and from the new mca param system
This commit was SVN r28741.
2013-07-09 14:37:08 +00:00
Ralph Castain
315da8125d Remove stale headers
cmr:v1.7.3:reviewer=jsquyres

This commit was SVN r28732.
2013-07-08 18:26:58 +00:00
Joshua Ladd
0b5c1f2ea8 Add 'generic' support for PMI2 (previously, we checked for PMI2 only on Cray systems.) If your resource manager (e.g. SLURM) has support for PMI2, then the --with-pmi configure flag will enable its usage. If you don't have PMI2, then you will fallback to regular old PMI1. This patch was submitted by Ralph Castain and reviewed and pushed by Josh Ladd. This should be added to cmr:v1.7:reviewer=jladd
This commit was SVN r28666.
2013-06-21 15:28:14 +00:00
Nathan Hjelm
518d1fe200 Fix two typos that prevented alps direct launch from working
This commit was SVN r28628.
2013-06-13 17:04:08 +00:00
Ralph Castain
45af6cf59e The move of the orte_db framework to opal required that we create an opaque opal_identifier_t type as OPAL cannot know anything about the ORTE process name. However, passing a value down to opal and then having the db components reference it causes alignment issues on Solaris Sparc platforms. So pass the pointer instead and do the old "memcpy" trick to avoid the problem.
This commit was SVN r28308.
2013-04-08 23:34:16 +00:00
Ralph Castain
4dbc468c3c Remove stale file
This commit was SVN r28299.
2013-04-07 13:52:48 +00:00
Ralph Castain
c121a784ae Remove some weird code around opal_db_close and cleanup that framework's open/close operation
This commit was SVN r28298.
2013-04-07 13:52:28 +00:00
Ralph Castain
10257b8b43 Add missing include
This commit was SVN r28297.
2013-04-07 01:32:08 +00:00
Ralph Castain
1067b1f5ee Add a little debug
This commit was SVN r28295.
2013-04-06 15:24:35 +00:00
Nathan Hjelm
365cf48db5 Update OPAL frameworks to use the MCA framework system.
This commit was SVN r28239.
2013-03-27 21:11:47 +00:00
Nathan Hjelm
cf377db823 MCA/base: Add new MCA variable system
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.
2013-03-27 21:09:41 +00:00
Ralph Castain
8d2fa3693b First cut at removing the native Windows support. Remove all the Windows-specific components, and the .windows files sprinkled around. Remove the Windows platform files and MTT scripts. Update the NEWS to point Windows users to the cygwin package.
This commit was SVN r28116.
2013-02-26 20:44:56 +00:00
Ralph Castain
9479635e31 Missing include here too...
This commit was SVN r28115.
2013-02-26 20:21:10 +00:00
Ralph Castain
8b8333da3e Add missing include
This commit was SVN r28114.
2013-02-26 19:56:05 +00:00
Ralph Castain
bd9265c560 Per the meeting on moving the BTLs to OPAL, move the ORTE database "db" framework to OPAL so the relocated BTLs can access it. Because the data is indexed by process, this requires that we define a new "opal_identifier_t" that corresponds to the orte_process_name_t struct. In order to support multiple run-times, this is defined in opal/mca/db/db_types.h as a uint64_t without identifying the meaning of any part of that data.
A few changes were required to support this move:

1. the PMI component used to identify rte-related data (e.g., host name, bind level) and package them as a unit to reduce the number of PMI keys. This code was moved up to the ORTE layer as the OPAL layer has no understanding of these concepts. In addition, the component locally stored data based on process jobid/vpid - this could no longer be supported (see below for the solution).

2. the hash component was updated to use the new opal_identifier_t instead of orte_process_name_t as its index for storing data in the hash tables. Previously, we did a hash on the vpid and stored the data in a 32-bit hash table. In the revised system, we don't see a separate "vpid" field - we only have a 64-bit opaque value. The orte_process_name_t hash turned out to do nothing useful, so we now store the data in a 64-bit hash table. Preliminary tests didn't show any identifiable change in behavior or performance, but we'll have to see if a move back to the 32-bit table is required at some later time.

3. the db framework was a "select one" system. However, since the PMI component could no longer use its internal storage system, the framework has now been changed to a "select many" mode of operation. This allows the hash component to handle all internal storage, while the PMI component only handles pushing/pulling things from the PMI system. This was something we had planned for some time - when fetching data, we first check internal storage to see if we already have it, and then automatically go to the global system to look for it if we don't. Accordingly, the framework was provided with a custom query function used during "select" that lets you seperately specify the "store" and "fetch" ordering.

4. the ORTE grpcomm and ess/pmi components, and the nidmap code,  were updated to work with the new db framework and to specify internal/global storage options.

No changes were made to the MPI layer, except for modifying the ORTE component of the OMPI/rte framework to support the new db framework.

This commit was SVN r28112.
2013-02-26 17:50:04 +00:00