the ofi mtl mrecv was not properly setting the message in/out
arg to MPI_MRECV to MPI_MESSAGE_NULL.
Signed-off-by: Howard Pritchard <hppritcha@gmail.com>
The btl/ofi does not currently utilize the common ofi include/exclude
list. Added verification code similar to the mtl/ofi that will check if
the info object is in the include or exclude list. If it isn't in the
include list or is in the exclude list, validate_info will return
OPAL_ERROR. The btl/ofi will no longer pass a provider name as a hint
when calling getinfo, instead filtering the provider during
validate_info.
This patch also moves the is_in_list MTL function into common code and
adds additional debugging output to the BTL to match the MTL standard.
Signed-off-by: William Zhang <wilzhang@amazon.com>
Once any number of events are read, return immediately, rather than
waiting for fi_cq_read() to return FI_EAGAIN or an error. This can
improve observed latency if the user application is in a blocking call
waiting for us to return. Deleting the while loop here also means
ofi_progress_event_count serves as an upper bound for the total number
of events read in a single call (with the while loop we might read far
more, as long as new events continue to arrive).
Signed-off-by: Eric Badger <eric@badgerio.us>
also add common verbose variable.
Note the verbosity thing is a little tricky owing to the way the MCA frameworks and components are registered and
and initialized. The BTL's are registered/initialized prior to the MTL components even getting registered.
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
Adds the capability to select a NIC based on hardware locality.
Creates a list of NICs that share the same cpuset as the process,
then selects the NIC based on the (local rank) % (number of NICs).
If no NICs are available that share the same cpuset, the selection process
will create a list of all available NICs and make a selection based on
(local rank) % (number of NICs)
Signed-off-by: Nikola Dancejic <dancejic@amazon.com>
Change ompi_mtl_ofi_get_endpoint() to call the active PML's
add_procs() rather than the OFI MTL add_procs() directly when
discovering a new process during operation.
Functionally, this has no impact in correct operation. However,
the current behavior means that the heterogenous and active PML
checks are not being executed in the dynamic discovery case.
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
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>
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>
Some versions of Libfabric contain a bug in EFA where FI_REMOTE_COMM and
FI_LOCAL_COMM are not advertised. In order to workaround this, we need to call
fi_getinfo() without those capability bits to see if EFA is available first.
Also move around some of the provider include/exclude list logic so we can skip
this workaround if applicable.
Signed-off-by: Robert Wespetal <wesper@amazon.com>
Make sure to get an RDM provider that can provide both local and
remote communication. We need this check because some providers could
be selected via RXD or RXM, but can't provide local communication, for
example.
Add OPAL_CHECK_OFI_VERSION_GE() m4 macro to check that the Libfabric
we're building against is >= a target version. Use this check in two
places:
1. MTL/OFI: Make sure it is >= v1.5, because the FI_LOCAL_COMM /
FI_REMOTE_COMM constants were introduced in Libfabric API v1.5.
2. BTL/usnic: It already had similar configury to check for Libfabric
>= v1.1, but the usnic component was checking for >= v1.3. So
update the btl/usnic configury to use the new macro and check for
>= v1.3.
Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
Change the provider include and exclude list name comparison check to
ignore case. The UDP provider's name is uppercase and was being selected
despite being in the exclude list.
Signed-off-by: Robert Wespetal <wesper@amazon.com>
one off patch for v4.0.x. for some reason commit on master
didn't have this problem.
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
(cherry picked from commit 5f3dbdb5c8a94a4f426ecca1a3a91c83035f956c)
Note that this commit is actually a cherry-pick from the v4.0.x
branch. This is the opposite direction than what we nornmally do: we
usually commit to master first and then cherry-pick to the release
branches (vs. the other way around).
As is probably evident from the original commit message above, through
a comedy of errors, this commit was actually applied to the v4.0.x
branch first and then cherry-picked back to master (i.e., the problem
*did* exist in the original master commit
3aca4af548a3d781b6b52f89f4d6c7e66d379609, but it was not recongized at
the time).
Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
INTERNAL: STL-59403
The OFI (libfabric) MTL does not respect the maximum message size
parameter that OFI provides in the fi_info data.
This patch adds this missing max_msg_size field to the mca_ofi_module_t
structure and adds a length check to the low-level send routines.
Change-Id: I05aa71d332f2df897133b30c28bf37d98f061996
Signed-off-by: Michael Heinz <michael.william.heinz@intel.com>
Reviewed-by: Adam Goldman <adam.goldman@intel.com>
Reviewed-by: Brendan Cunningham <brendan.cunningham@intel.com>
Updated the OFI MTL's Recv cancel to be a non-blocking call to match
the MPI spec. Given fi_cancel succeeded, then it is expected that the
user will wait on the request to read the result of if the cancel has
completed.
Signed-off-by: Spruit, Neil R <neil.r.spruit@intel.com
For the non thread-grouping paths, only the first (0th) OFI context
should be used for communication. Otherwise this would access a non existant
array item and cause segfault.
While at it, clarifiy some content regarding SEPs in README (Credit to Matias Cabral
for README edits).
Signed-off-by: Aravind Gopalakrishnan <Aravind.Gopalakrishnan@intel.com>
Update the OPAL_CHECK_OFI configury macro:
- Make it safe to call the macro multiple times:
- The checks only execute the first time it is invoked
- Subsequent invocations, it just emits a friendly "checking..."
message so that configure output is sensible/logical
- With the goal of ultimately removing opal/mca/common/ofi, rename the
output variables from OPAL_CHECK_OFI to be
opal_ofi_{happy|CPPFLAGS|LDFLAGS|LIBS}.
- Update btl/ofi, btl/usnic, and mtl/ofi for these new conventions.
- Also, don't use AC_REQUIRE to invoke OPAL_CHECK_OFI because that
causes the macro to be invoked at a fairly random time, which makes
configure stdout confusing / hard to grok.
- Remove a little left-over kruft in OPAL_CHECK_OFI, too (which
resulted in an indenting change, making the change to
opal_check_ofi.m4 look larger than it really is).
Thanks Alastair McKinstry for the report and initial fix.
Thanks Rashika Kheria for the reminder.
Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
When we exceed the threshold number of contexts created, print appropriate help
text
Signed-off-by: Aravind Gopalakrishnan <Aravind.Gopalakrishnan@intel.com>
Provide the av_attr.count hint (number of addresses that will be
inserted into the address vector through the life of the process)
at initialization of the address vector. It's ok to be a bit
wrong, but some endpoints (RxR) can benefit by not going through
the slow growth realloc churn.
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
With MTLs, there's no "other transport" when the remote side
does not have an active NIC, so we should print a useful error
message when the modex failed (indicating lack of a NIC on
the remote side).
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
Moving to a model where we have users actively _enable_ SEP feature for use
rather than opening SEP by default if provider supports it. This allows us to
not regress (either functionally or for performance reasons) any apps that were
working correctly on regular endpoints.
Also, providing MCA to specify number of OFI contexts to create and default
this value to 1 (Given btl/ofi also creates one by default, this reduces the
incidence of a scenario where we allocate all available contexts by default and
if btl/ofi asks for one more, then provider breaks as it doesn't support it).
While at it, spruce up README on SEP content.
Signed-off-by: Aravind Gopalakrishnan <Aravind.Gopalakrishnan@intel.com>
-> Added new targets in Makefile.am to call a new build script
generate-opt-funcs.pl to generate specialized functions for
each *.pm file.
-> Added new perl module *.pm files for send,isend,irecv,iprobe,improbe
which are loaded by generate-opt-funcs.pl to create new source files
that correspond to the name of the .pm file to be used as part of
MTL OFI.
-> Added mtl_ofi_opt.pm.template and updated README with details on the
specialization features and how to add additional specialization
support.
-> Added new opt_common/mtl_ofi_opt_common.pm containing common
functions for generating the specialized functions used by
all other *.pm modules.
-> Added new mtl_ofi.h which includes the definitions for the
function symbol table for storing the specialized functions along
with the definitions for the initialization functions for the
corresponding function pointers.
-> Based off the OFI provider capabilities the specialized function
pointers are assigned at mtl_ofi_component_init to the corresponding
MTL OFI function.
-> mca_mtl_ofi_module_t has been updated with the symbol table
struct which is assigned at component init.
Signed-off-by: Spruit, Neil R <neil.r.spruit@intel.com>
For cases when the number of local processes is greater than the number of
available contexts, the SEP initialization phase would calculate the number of
contexts to provision for each rank to be 0 and would eventually crash.
Fix the issue here by using regular endpoints in the event the number of local
processes is more than available contexts. This fixes issue #6182.
Signed-off-by: Aravind Gopalakrishnan <Aravind.Gopalakrishnan@intel.com>
Commit 109d0569ffd introduced a crash when an error occurred
before ofi_ctxt was allocated, including when no providers
passed the selection logic. Properly check that the pointer
is not NULL in the error cleanup code before dereferencing
the pointer.
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
OFI MTL supports OFI Scalable Endpoints feature as means to improve
multi-threaded application throughput and message rate. Currently the feature
is designed to utilize multiple TX/RX contexts exposed by the OFI provider in
conjunction with a multi-communicator MPI application model. For more
information, refer to README under mtl/ofi.
Reviewed-by: Matias Cabral <matias.a.cabral@intel.com>
Reviewed-by: Neil Spruit <neil.r.spruit@intel.com>
Signed-off-by: Aravind Gopalakrishnan <Aravind.Gopalakrishnan@intel.com>
So far Vader is faster than OFI MTL for doing shared memory.
Therefore, let it run by default when all procs are local.
Reviewed-by: Spruit, Neil R <neil.r.spruit@intel.com>
Reviewed-by: Gopalakrishnan, Aravind <aravind.gopalakrishnan@intel.com>
Signed-off-by: Matias Cabral <matias.a.cabral@intel.com>
When the provider does not support FI_REMOTE_CQ_DATA, the OFI tag does not have
sizeof(int) bits for the rank. Therefore, unexpected behavior will occur when
this limit is crossed.
Check the max allowed number of ranks during add_procs() and return if there is
danger of exceeding this threshold.
Signed-off-by: Aravind Gopalakrishnan <Aravind.Gopalakrishnan@intel.com>
The Open MPI code base assumed that asprintf always behaved like
the FreeBSD variant, where ptr is set to NULL on error. However,
the C standard (and Linux) only guarantee that the return code will
be -1 on error and leave ptr undefined. Rather than fix all the
usage in the code, we use opal_asprintf() wrapper instead, which
guarantees the BSD-like behavior of ptr always being set to NULL.
In addition to being correct, this will fix many, many warnings
in the Open MPI code base.
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
Change default provider selection logic for the OFI MTL. The
old logic was whitelist-only, so any new HPC NIC provider would
have to ask users to do extra work or wait for an OMPI release
to be whitelisted. The reason for the logic was to avoid
selecting a "generic" provider like sockets or shm that would
frequently have worse performance than the optimized BTL options
Open MPI supports.
With the change, we blacklist the (small, relatively static) list
of providers that duplicate internal capabilities. Users can use
one of thse blacklisted providers in two ways: first, they can
explicitly request the provider in the include list (which will
override the default exclude list) and second, the can set a new
empty exclude list.
Since most HPC networks require special libraries and therefore
an explicit build of libfabric, it is highly unlikely that this
change will cause users to use libfabric when they didn't want to
do so. It does, however, solve the whitelisting problem.
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
When an application is not using multiple threads to call into MPI, we can
safely ask for FI_THREAD_DOMAIN setting from the provider as it should
translate to the least amount of locking in provider.
Conversely, for applications using THREAD_MULTIPLE, explicitly ask for
FI_THREAD_SAFE to prevent race conditions.
Signed-off-by: Aravind Gopalakrishnan <Aravind.Gopalakrishnan@intel.com>
Since progress entries array is globally allocated, it is susceptible
to race conditions when using multi-threaded applications. Allocating it
on the stack resolves any potential races as it is thread local by default.
Signed-off-by: Aravind Gopalakrishnan <Aravind.Gopalakrishnan@intel.com>
-Updated blocking send to directly call functionality and
set completion events expected to 0 initally. This allows for optimization for
providers that support fi_tinject up to larger sizes. This also reduces
latency on running the OFI mtl with smaller sizes without requiring
calls to progress given fi_tinject is required to complete the messaging
before returning and will not create any events in the Completion Queue.
-Updated non-blocking send to directly call fi_tsend and avoid calling
fi_tinject as the functionality should not wait on completions. This
resolves a bug where applications calling MPI_Isend can overrun the
TX buffer with small (inject) messages causing a deadlock. In addition
this improves performance in message rates by preventing
waiting on any size message to complete in non-blocking send messages.
-Created common ompi_mtl_ofi_ssend_recv function to post the ssend recv
which is common between isend and send code paths.
Signed-off-by: Spruit, Neil R <neil.r.spruit@intel.com>
- If a message for a recv that is being cancelled gets completed after
the call to fi_cancel, then the OFI mtl will enter a deadlock state
waiting for ofi_req->super.ompi_req->req_status._cancelled which will
never happen since the recv was successfully finished.
- To resolve this issue, the OFI mtl now checks ofi_req->req_started
to see if the request has been started within the loop waiting for the
event to be cancelled. If the request is being completed, then the loop
is broken and fi_cancel exits setting
ofi_req->super.ompi_req->req_status._cancelled = false;
Signed-off-by: Spruit, Neil R <neil.r.spruit@intel.com>
OFI providers may reserve some of the upper bits of the tag for
internal usage and expose it using mem_tag_format. Check for that
and adjust communicator bits as needed.
Signed-off-by: Aravind Gopalakrishnan <Aravind.Gopalakrishnan@intel.com>
- Added support in MTL_OFI_RETRY_UNTIL_DONE to handle -FI_EAGAIN
from the provider and correctly attempt to progress the OFI Completion
queue by calling ompi_mtl_ofi_progress.
- If events were pending that blocked OFI operations from being enqueued
they will be completed and the OFI operation will be retried once
ompi_mtl_ofi_progress has successfully completed.
- Updated MTL_OFI_RETRY_UNTIL_DONE to take a RETURN variable instead of
requiring the existance of a "ret" variable to pass back the return
value from completing the OFI operation.
Signed-off-by: Spruit, Neil R <neil.r.spruit@intel.com>
-Updated the design for sync send MPI calls to use 2 protocol bits for
denoting "sync_send" or "sync_send_ack".
-"Sync_send" is added to the send tag only and is masked out in receives
such that it can be read by the original Recv posted in the send/recv
operation.
-"Sync_send_ack" is sent from the recv callback to the send side. This 0
byte send does not generate a completion entry and instead sends the
message and immediately completes the opal completion in the recv.
-Tag formats ofi_tag_1 and ofi_tag_2 have been updated to include 2
more tag bits per format type due to the reduced protocal bits required
by OMPI.
Signed-off-by: Spruit, Neil R <neil.r.spruit@intel.com>
Extend number of supported ranks with providers that support
FI_REMOTE_CQ_DATA. Add README file to OFI MTL
Signed-off-by: Matias Cabral <matias.a.cabral@intel.com>
-Updated ompi_mtl_ofi_progress to use an array to read CQ events up to a
threshold that can be set by the Open MPI User.
-Users can adjust the number of events that can be handled in the
ompi_mtl_ofi_progress by setting "--mca mtl_ofi_progress_event_cnt #".
-The default value for the the number of CQ events that can be read in a
single call to ofi progress is 100 which is an average
based off workload usecase anaylsis showing 70-128 as the range of
multiple events returned during ofi progress.
Signed-off-by: Spruit, Neil R <neil.r.spruit@intel.com>
This fixes a regression in sockets provider which could return -EINTR value
from fi_cq_read() due to a syscall being interrupted. The error value is
currently interpreted as fatal condition. Relax the rule so that we can retry
fi_cq_read() operation.
Signed-off-by: Aravind Gopalakrishnan <Aravind.Gopalakrishnan@intel.com>
At least with some providers (sockets and GNI), the mprobe/mrecv
ofi mtl methods were incorrect. For these two providers at least
one must supply the original tag and mask bits used with the
prior FI_PEEK | FI_CLAIM request that had been used to probe for
the message.
These providers take a strict interpretation of the following sentence
from the libfabric fi_tagged man page:
```
Claimed messages can only be retrieved using a subsequent, paired receive operation with the FI_CLAIM flag set.
```
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
gcc 5.2 complains:
```
mtl_ofi_component.c: In function ‘ompi_mtl_ofi_finalize’:
mtl_ofi_component.c:613:5: warning: suggest parentheses around assignment used as truth value [-Wparentheses]
if (ret = fi_close((fid_t)ompi_mtl_ofi.fabric)) {
^
```
Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
