ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
9bc7a54e0c
openmpi/opal/mca/btl/usnic/btl_usnic.h

263 строки
7.8 KiB
C
Исходник Обычный вид История

First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
/*
* Copyright (c) 2004-2008 The Trustees of Indiana University and Indiana
* University Research and Technology
* Corporation. All rights reserved.
* Copyright (c) 2004-2011 The University of Tennessee and The University
* of Tennessee Research Foundation. All rights
* reserved.
* Copyright (c) 2004-2005 High Performance Computing Center Stuttgart,
* University of Stuttgart. All rights reserved.
* Copyright (c) 2004-2005 The Regents of the University of California.
* All rights reserved.
* Copyright (c) 2006 Sandia National Laboratories. All rights
* reserved.
usnic: fix fi_av_insert (ARP resolution) bugs We had several problems in the old code: 1. We were specifying an arbitrary timeout (100 ms) and then abandoning all remaining pending AV insert operations. We would then free the endpoint buffer that we gave to fi_av_insert(), usually causing libfabric's progress thread to write to a freed buffer. 2. We were claiming in a show_help message that the timeout was controllable via an MCA parameter. This commit removes that parameter, since there's no good method for us to specify a timeout like this to libfabric right now. 3. We also weren't waiting for the correct number of fi_av_insert() operations to complete. We were waiting for nprocs, which is accidentally fine for 2 procs on separate hosts, but not for most other proc counts. Reviewed-by: Jeff Squyres <jsquyres@cisco.com>
2015-01-07 19:25:17 +03:00
* Copyright (c) 2011-2015 Cisco Systems, Inc. All rights reserved.
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
/**
* @file
*/
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) 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.
2014-07-26 04:47:28 +04:00
#ifndef OPAL_BTL_USNIC_H
#define OPAL_BTL_USNIC_H
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) 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.
2014-07-26 04:47:28 +04:00
#include "opal_config.h"
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
#include <sys/types.h>
usnic: stomp format specifier warnings The usnic BTL now builds cleanly under `--enable-picky` when `MSGDEBUG1` is set. Reviewed-by: jsquyres cmr=v1.7.4:reviewer=jsquyres This commit was SVN r29097.
2013-08-30 03:24:14 +04:00
#include "opal_stdint.h"
Per the RFC schedule, add an additive lagged Fibonacci parallel random number generator to OPAL. In order to use, please add the following header to your code: opal/util/alfg.h. See ompi/mca/btl/openib/connect/btl_openib_connect_udcm.c for an example how to seed with opal_srand and invoke the generator with opal_rand. This should be added to cmr=v1.7.5:reviewer=rhc:subject=Add an OPAL RNG This commit was SVN r30801.
2014-02-24 01:41:38 +04:00
#include "opal/util/alfg.h"
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
#include "opal/class/opal_hash_table.h"
#include "opal/class/opal_hash_table.h"
#include "opal/mca/event/event.h"
usnic: update to new mpool API NOTE: Have not added cross-compatibility with v1.8 branch yet
2015-02-23 19:07:45 +03:00
#if BTL_IN_OPAL
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) 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.
2014-07-26 04:47:28 +04:00
#include "opal/mca/btl/btl.h"
#include "opal/mca/btl/base/btl_base_error.h"
#include "opal/mca/btl/base/base.h"
#include "opal/mca/mpool/grdma/mpool_grdma.h"
usnic: convert from verbs to libfabric (yay!) This commit represents the conversion of the usnic BTL from verbs to libfabric. For the moment, libfabric is embedded in Open MPI (currently in the usnic BTL). This is because the libfabric API is still changing, and also has not yet been released. Ultimately, this embedded copy of libfabric will likely disappear and the usnic BTL will rely on an external installation of libfabric. New configure options: * --with-libfabric: will cause configure to fail if libfabric support cannot be built * --without-libfabric: will prevent libfabric support from being built * --with-libfabric=DIR: use an external libfabric installation * --with-libfabric-libdir=LIBDIR: when paired with --with-libfabric=DIR, use LIBDIR for the libfabric installation library dir The --with-libnl3[-libdir] arguments are now gone.
2014-12-03 00:09:46 +03:00
#else
#include "ompi/mca/btl/btl.h"
#include "ompi/mca/btl/base/btl_base_error.h"
#include "ompi/mca/btl/base/base.h"
#include "ompi/mca/mpool/grdma/mpool_grdma.h"
#endif
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
usnic: v1.6<->trunk unification, trunk side The Cisco-maintained v1.6 port of the usnic BTL has diverged from the upstream trunk and v1.7 branches. This commit adjusts the trunk to more closely match the v1.6 branch to simplify future merging and cherry-picking. The usnic MCA parameters also need work on this side. Should be included in usnic v1.7.3 roll-up CMR (refs trac:3760) This commit was SVN r29138. The following Trac tickets were found above: Ticket 3760 --> https://svn.open-mpi.org/trac/ompi/ticket/3760
2013-09-06 07:21:34 +04:00
#include "btl_usnic_compat.h"
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
BEGIN_C_DECLS
/*
* We're simulating a clock as best we can without resorting to the
* system. The clock is used to defer ACKs, and ticks will be incremented
* when progression gets called. It could be incremented by different amounts
* at other times as needed or as tuning dictates.
*/
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) 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.
2014-07-26 04:47:28 +04:00
extern uint64_t opal_btl_usnic_ticks;
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
static inline uint64_t
get_nsec(void)
{
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) 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.
2014-07-26 04:47:28 +04:00
return opal_btl_usnic_ticks;
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
}
Per the RFC schedule, add an additive lagged Fibonacci parallel random number generator to OPAL. In order to use, please add the following header to your code: opal/util/alfg.h. See ompi/mca/btl/openib/connect/btl_openib_connect_udcm.c for an example how to seed with opal_srand and invoke the generator with opal_rand. This should be added to cmr=v1.7.5:reviewer=rhc:subject=Add an OPAL RNG This commit was SVN r30801.
2014-02-24 01:41:38 +04:00
/* RNG buffer declaration */
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) 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.
2014-07-26 04:47:28 +04:00
extern opal_rng_buff_t opal_btl_usnic_rand_buff;
Per the RFC schedule, add an additive lagged Fibonacci parallel random number generator to OPAL. In order to use, please add the following header to your code: opal/util/alfg.h. See ompi/mca/btl/openib/connect/btl_openib_connect_udcm.c for an example how to seed with opal_srand and invoke the generator with opal_rand. This should be added to cmr=v1.7.5:reviewer=rhc:subject=Add an OPAL RNG This commit was SVN r30801.
2014-02-24 01:41:38 +04:00
Protect the container_of #define. MOFED apparently has a /usr/include/infiniband/verbs.h that also defines a (slightly different but fully compatible) container_of macro. So put proper #ifndef protection around our definition of container_of. Thanks to Rolf vandeVaart for pointing out the issue. Reviewed by Dave Goodell. cmr=v1.7.4:reviewer=ompi-rm1.7 This commit was SVN r29799.
2013-12-04 18:24:56 +04:00
#ifndef container_of
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
#define container_of(ptr, type, member) ( \
(type *)( ((char *)(ptr)) - offsetof(type,member) ))
Protect the container_of #define. MOFED apparently has a /usr/include/infiniband/verbs.h that also defines a (slightly different but fully compatible) container_of macro. So put proper #ifndef protection around our definition of container_of. Thanks to Rolf vandeVaart for pointing out the issue. Reviewed by Dave Goodell. cmr=v1.7.4:reviewer=ompi-rm1.7 This commit was SVN r29799.
2013-12-04 18:24:56 +04:00
#endif
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
usnic: convert from verbs to libfabric (yay!) This commit represents the conversion of the usnic BTL from verbs to libfabric. For the moment, libfabric is embedded in Open MPI (currently in the usnic BTL). This is because the libfabric API is still changing, and also has not yet been released. Ultimately, this embedded copy of libfabric will likely disappear and the usnic BTL will rely on an external installation of libfabric. New configure options: * --with-libfabric: will cause configure to fail if libfabric support cannot be built * --without-libfabric: will prevent libfabric support from being built * --with-libfabric=DIR: use an external libfabric installation * --with-libfabric-libdir=LIBDIR: when paired with --with-libfabric=DIR, use LIBDIR for the libfabric installation library dir The --with-libnl3[-libdir] arguments are now gone.
2014-12-03 00:09:46 +03:00
#ifndef max
#define max(a, b) (((a) > (b)) ? (a) : (b))
usnic: handle missing ibv_event_type_str Some older versions of libibverbs do not have `ibv_event_type_str`, leading to compilation failures on older machines, irrespective of whether they could ever support usNIC anyway. If we encounter any other build issues related to "old verbs" then we should just cause the usnic BTL to disqualify itself when it encounters "old" traits. Thanks to Paul Hargrove for reporting the issue: http://www.open-mpi.org/community/lists/devel/2014/02/14056.php Reviewed-by: Jeff Squyres <jsquyres@cisco.com> cmr=v1.7.5:reviewer=ompi-rm1.7 This commit was SVN r30674.
2014-02-11 23:18:29 +04:00
#endif
Clean up debugging logging a bit. MSGDEBUG2 now means "print a one-liner for all PML calls into BTL, and also when BTL calls PML with a recv completion (not send completions)" MSGDEBUG1 means print more internal gory detail MSGDEBUG is gone, replaced by MSGDEBUG1 In the process also found that PUT_DEST style fragments could potentially be leaked in usnic_free() since send_fragment tests were being applied to see if it was eligible to be freed. This commit was SVN r29185.
2013-09-17 11:29:40 +04:00
/* MSGDEBUG2 prints 1 line at each BTL entry point */
#define MSGDEBUG2 (MSGDEBUG1||0)
/* MSGDEBUG1 prints more info about arguments and internal functions */
#define MSGDEBUG1 0
usnic: add MSGDEBUG1_OUT/MSGDEBUG2_OUT macros This includes suppressing picky-mode warnings about __VA_ARGS__, which we know are supported by any compilers we care about. Reviewed-by: Jeff Squyres <jsquyres@cisco.com> Reviewed-by: Reese Faucette <rfaucett@cisco.com> This commit was SVN r29590.
2013-11-05 02:51:35 +04:00
/* output macros to declutter source */
#if MSGDEBUG1
#define MSGDEBUG1_OUT(...) opal_output(0, __VA_ARGS__)
#else
#define MSGDEBUG1_OUT(...) do {} while (0)
#endif
#if MSGDEBUG2
#define MSGDEBUG2_OUT(...) opal_output(0, __VA_ARGS__)
#else
#define MSGDEBUG2_OUT(...) do {} while (0)
#endif
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
/* Set to >0 to randomly drop received frags. The higher the number,
the more frequent the drops. */
#define WANT_RECV_FRAG_DROPS 0
/* Set to >0 to randomly fail to send an ACK, mimicing a lost ACK.
The higher the number, the more frequent the failed-to-send-ACK. */
#define WANT_FAIL_TO_SEND_ACK 0
/* Set to >0 to randomly fail to resend a frag (causing it to be
requed to be sent later). The higher the number, the more frequent
the failed-to-resend-frag. */
#define WANT_FAIL_TO_RESEND_FRAG 0
#if WANT_RECV_FRAG_DROPS > 0
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) 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.
2014-07-26 04:47:28 +04:00
#define FAKE_RECV_FRAG_DROP (opal_rand(&opal_btl_usnic_rand_buff) < WANT_RECV_FRAG_DROPS)
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
#else
#define FAKE_RECV_FRAG_DROP 0
#endif
#if WANT_FAIL_TO_SEND_ACK > 0
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) 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.
2014-07-26 04:47:28 +04:00
#define FAKE_FAIL_TO_SEND_ACK (opal_rand(&opal_btl_usnic_rand_buff) < WANT_FAIL_TO_SEND_ACK)
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
#else
#define FAKE_FAIL_TO_SEND_ACK 0
#endif
#if WANT_FAIL_TO_RESEND_FRAG > 0
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) 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.
2014-07-26 04:47:28 +04:00
#define FAKE_FAIL_TO_RESEND_FRAG (opal_rand(&opal_btl_usnic_rand_buff) < WANT_FAIL_TO_RESEND_FRAG)
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
#else
#define FAKE_FAIL_TO_RESEND_FRAG 0
#endif
/**
usnic: convert from verbs to libfabric (yay!) This commit represents the conversion of the usnic BTL from verbs to libfabric. For the moment, libfabric is embedded in Open MPI (currently in the usnic BTL). This is because the libfabric API is still changing, and also has not yet been released. Ultimately, this embedded copy of libfabric will likely disappear and the usnic BTL will rely on an external installation of libfabric. New configure options: * --with-libfabric: will cause configure to fail if libfabric support cannot be built * --without-libfabric: will prevent libfabric support from being built * --with-libfabric=DIR: use an external libfabric installation * --with-libfabric-libdir=LIBDIR: when paired with --with-libfabric=DIR, use LIBDIR for the libfabric installation library dir The --with-libnl3[-libdir] arguments are now gone.
2014-12-03 00:09:46 +03:00
* usnic BTL component
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
*/
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) 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.
2014-07-26 04:47:28 +04:00
typedef struct opal_btl_usnic_component_t {
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
/** base BTL component */
mca_btl_base_component_2_0_0_t super;
usnic: rearrange component struct field order Just trying to be deliberate about keeping fastpath-accessed fields grouped together to fit into the same 64-byte cache line. Reviewed-by: Jeff Squyres <jsquyres@cisco.com> cmr=v1.7.5:ticket=trac:4253 This commit was SVN r30837. The following Trac tickets were found above: Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
2014-02-26 11:40:53 +04:00
/* in the v1.6 series, sizeof(super) is 256, leading to good alignment for
* subsequent fastpath fields */
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
/** Maximum number of BTL modules */
usnic: convert from verbs to libfabric (yay!) This commit represents the conversion of the usnic BTL from verbs to libfabric. For the moment, libfabric is embedded in Open MPI (currently in the usnic BTL). This is because the libfabric API is still changing, and also has not yet been released. Ultimately, this embedded copy of libfabric will likely disappear and the usnic BTL will rely on an external installation of libfabric. New configure options: * --with-libfabric: will cause configure to fail if libfabric support cannot be built * --without-libfabric: will prevent libfabric support from being built * --with-libfabric=DIR: use an external libfabric installation * --with-libfabric-libdir=LIBDIR: when paired with --with-libfabric=DIR, use LIBDIR for the libfabric installation library dir The --with-libnl3[-libdir] arguments are now gone.
2014-12-03 00:09:46 +03:00
int max_modules;
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
/** Number of available/initialized BTL modules */
usnic: convert from verbs to libfabric (yay!) This commit represents the conversion of the usnic BTL from verbs to libfabric. For the moment, libfabric is embedded in Open MPI (currently in the usnic BTL). This is because the libfabric API is still changing, and also has not yet been released. Ultimately, this embedded copy of libfabric will likely disappear and the usnic BTL will rely on an external installation of libfabric. New configure options: * --with-libfabric: will cause configure to fail if libfabric support cannot be built * --without-libfabric: will prevent libfabric support from being built * --with-libfabric=DIR: use an external libfabric installation * --with-libfabric-libdir=LIBDIR: when paired with --with-libfabric=DIR, use LIBDIR for the libfabric installation library dir The --with-libnl3[-libdir] arguments are now gone.
2014-12-03 00:09:46 +03:00
int num_modules;
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
Use the RTE framework instead of calling ORTE directly. Brian (rightfully) hit me on the head with the don't-use-ORTE-use-the-rte-framework clue bat; the usnic BTL now nicely plays with the RTE framework. This commit was SVN r28907.
2013-07-22 21:28:23 +04:00
/* Cached hashed version of my RTE proc name (to stuff in
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
protocol headers) */
usnic: convert from verbs to libfabric (yay!) This commit represents the conversion of the usnic BTL from verbs to libfabric. For the moment, libfabric is embedded in Open MPI (currently in the usnic BTL). This is because the libfabric API is still changing, and also has not yet been released. Ultimately, this embedded copy of libfabric will likely disappear and the usnic BTL will rely on an external installation of libfabric. New configure options: * --with-libfabric: will cause configure to fail if libfabric support cannot be built * --without-libfabric: will prevent libfabric support from being built * --with-libfabric=DIR: use an external libfabric installation * --with-libfabric-libdir=LIBDIR: when paired with --with-libfabric=DIR, use LIBDIR for the libfabric installation library dir The --with-libnl3[-libdir] arguments are now gone.
2014-12-03 00:09:46 +03:00
uint64_t my_hashed_rte_name;
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
Fix a collection of bugs found by QA and Coverity, and make some minor improvements: * Fix minor memory leaks during component_init * Ensure that an initialization loop does not underflow an unsigned int * Improve mlock limit checking * Fix set of BTL modules created during component_init when failing to get QP resources or otherwise excluding some (but not all) usnic verbs devices * Fix/improve error messages to be consistent with other Cisco documentation * Randomize the initial sliding window sequence number so that we silently drop incoming frames from previous jobs that still have existant processes in the middle of dying (and are still transmitting) * Ensure we don't break out of add_procs too soon and create an asymetrical view of what interfaces are available This commit was SVN r28975.
2013-08-01 20:56:15 +04:00
/** array of possible BTLs (>= num_modules elements) */
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) 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.
2014-07-26 04:47:28 +04:00
struct opal_btl_usnic_module_t* usnic_all_modules;
Fix a collection of bugs found by QA and Coverity, and make some minor improvements: * Fix minor memory leaks during component_init * Ensure that an initialization loop does not underflow an unsigned int * Improve mlock limit checking * Fix set of BTL modules created during component_init when failing to get QP resources or otherwise excluding some (but not all) usnic verbs devices * Fix/improve error messages to be consistent with other Cisco documentation * Randomize the initial sliding window sequence number so that we silently drop incoming frames from previous jobs that still have existant processes in the middle of dying (and are still transmitting) * Ensure we don't break out of add_procs too soon and create an asymetrical view of what interfaces are available This commit was SVN r28975.
2013-08-01 20:56:15 +04:00
/** array of pointers to active BTLs (num_modules elements) */
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) 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.
2014-07-26 04:47:28 +04:00
struct opal_btl_usnic_module_t** usnic_active_modules;
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
usnic: rearrange component struct field order Just trying to be deliberate about keeping fastpath-accessed fields grouped together to fit into the same 64-byte cache line. Reviewed-by: Jeff Squyres <jsquyres@cisco.com> cmr=v1.7.5:ticket=trac:4253 This commit was SVN r30837. The following Trac tickets were found above: Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
2014-02-26 11:40:53 +04:00
/** convertor packing threshold */
int pack_lazy_threshold;
/* vvvvvvvvvv non-fastpath fields go below vvvvvvvvvv */
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
/** list of usnic proc structures */
opal_list_t usnic_procs;
/** name of memory pool */
char* usnic_mpool_name;
usnic: rearrange component struct field order Just trying to be deliberate about keeping fastpath-accessed fields grouped together to fit into the same 64-byte cache line. Reviewed-by: Jeff Squyres <jsquyres@cisco.com> cmr=v1.7.5:ticket=trac:4253 This commit was SVN r30837. The following Trac tickets were found above: Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
2014-02-26 11:40:53 +04:00
char *if_include;
char *if_exclude;
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
/** Want stats? */
bool stats_enabled;
bool stats_relative;
int stats_frequency;
/** Whether we want to use NUMA distances to choose which usNIC
devices to use for short messages */
bool want_numa_device_assignment;
/** max send descriptors to post per module */
int32_t sd_num;
/** max receive descriptors per module */
int32_t rd_num;
/** max send/receive desriptors for priority channel */
int32_t prio_sd_num;
int32_t prio_rd_num;
/** max completion queue entries per module */
int32_t cq_num;
/** retrans characteristics */
int retrans_timeout;
usnic: improve interface matching (esp. for UDP) Prior to this commit we matched local interfaces to remote interfaces in order to create endpoints in a simplistic way. If any remote interfaces were on the same subnet as any of our local interfaces then only local interfaces would be paired (IP-routed remote interfaces would be ignored). This commit introduces a more general scheme which attempts to make the "best" pairing of local interfaces to remote interfaces. We now cast the problem as a graph theory problem known as the "Assignment Problem", or finding a maximum-cardinality, minimum-weight bipartite matching. We solve this problem by reducing the bipartite graph of interface connectivity to a flow network and then solving for a minimum cost flow. This is then easily converted into back into a matching on the original bipartite graph. In the new scheme, interfaces on the same subnet are preferred over interfaces requiring intermediate routing hops and higher bandwidth links are preferred over lower bandwidth links. Reviewed-by: Jeff Squyres <jsquyres@cisco.com> cmr=v1.7.5:ticket=trac:4253 This commit was SVN r30849. The following Trac tickets were found above: Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
2014-02-26 11:50:26 +04:00
usnic: convert from verbs to libfabric (yay!) This commit represents the conversion of the usnic BTL from verbs to libfabric. For the moment, libfabric is embedded in Open MPI (currently in the usnic BTL). This is because the libfabric API is still changing, and also has not yet been released. Ultimately, this embedded copy of libfabric will likely disappear and the usnic BTL will rely on an external installation of libfabric. New configure options: * --with-libfabric: will cause configure to fail if libfabric support cannot be built * --without-libfabric: will prevent libfabric support from being built * --with-libfabric=DIR: use an external libfabric installation * --with-libfabric-libdir=LIBDIR: when paired with --with-libfabric=DIR, use LIBDIR for the libfabric installation library dir The --with-libnl3[-libdir] arguments are now gone.
2014-12-03 00:09:46 +03:00
/** transport header length for all usNIC devices on this server
(it is guaranteed that all usNIC devices on a single server
will have the same underlying transport, and therefore the
same transport header length) */
int transport_header_len;
uint32_t transport_protocol;
/* what UDP port do we want to use? If 0, the system will pick.
If nonzero, it is used as the base -- the final number will be
(base+my_local_rank). */
int udp_port_base;
Add usnic connectivity-checking agent service. Basically: since usnic is a connectionless transport, we do not get OS-provided services "for free" that connection-oriented transports get, namely: "hey, I wasn't able to make a connection to peer X", and "hey, your connection to peer X has died." This connectivity-checker runs in a separate progress thread in the usnic BTL in local rank 0 on each server. Upon first send in any process, the connectivty-checker agent will send some UDP pings to the peer to ensure that we can reach it. If we can't, we'll abort the job with a nice show_help message. There's a lengthy comment in btl_usnic_connectivity.h explains the scheme and how it works. Reviewed by Dave Goodell. cmr=v1.7.5:ticket=trac:4253 This commit was SVN r30860. The following Trac tickets were found above: Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
2014-02-27 02:21:25 +04:00
usnic: behave better when ibv_create_ah() fails When ibv_create_ah() fails due to an address resolution failure, it really only means that we can't reach that one peer -- so we should just ignore that one peer. If ibv_create_ah() fails for some other reason, then give up on the entire usnic_X device. Change the show_help() message that is displayed when ibv_create_ah() fails due to address resolution failure; indicate that it's likely a routing problem. Also opal_output_verbose() the same info, since show_help() is de-duplicated (and this particular show_help() message can be squelched). Fixes Cisco bugs CSCup35851 and CSCup35872. cmr=v1.8.2:ticket=trac:4734 This commit was SVN r32061. The following Trac tickets were found above: Ticket 4734 --> https://svn.open-mpi.org/trac/ompi/ticket/4734
2014-06-21 00:53:50 +04:00
/** disable the "cannot find route" warnings (for network setups
where this is known/acceptable) */
bool show_route_failures;
Add usnic connectivity-checking agent service. Basically: since usnic is a connectionless transport, we do not get OS-provided services "for free" that connection-oriented transports get, namely: "hey, I wasn't able to make a connection to peer X", and "hey, your connection to peer X has died." This connectivity-checker runs in a separate progress thread in the usnic BTL in local rank 0 on each server. Upon first send in any process, the connectivty-checker agent will send some UDP pings to the peer to ensure that we can reach it. If we can't, we'll abort the job with a nice show_help message. There's a lengthy comment in btl_usnic_connectivity.h explains the scheme and how it works. Reviewed by Dave Goodell. cmr=v1.7.5:ticket=trac:4253 This commit was SVN r30860. The following Trac tickets were found above: Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
2014-02-27 02:21:25 +04:00
/** connectivity verification: ACK timeout, number of retries
before issue an error/abort the job */
bool connectivity_enabled;
int connectivity_ack_timeout;
int connectivity_num_retries;
This is what r30852 should have been: Consolidate into a single, outter loop of ibv_create_ah() calls Follow on to SVN trunk r30850: consolidate the ibv_create_ah() calls into a single loop, MPI_WAITALL-style. That is, call the (effectively non-blocking) ibv_create_ah() for each endpoint. If we get NULL+EAGAIN, it means that the UDP ARP is still ongoing down in the kernel, so just try again later. We put these all into a single loop because it allows us to parallelize the ARP progress in the kernel. cmr=v1.7.5:ticket=trac:4253 This commit was SVN r30879. The following SVN revision numbers were found above: r30850 --> open-mpi/ompi@3641500442c1362e148d0d6272db91d75ec92116 r30852 --> open-mpi/ompi@4e282a329578c9e7296d5a3809fedfa9f83870df The following Trac tickets were found above: Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
2014-02-27 21:19:50 +04:00
usnic: convert some BTL_ERRORs to more descriptive show_help messages 1. After we receive N abnormally-short messages (meaning: corrupted), print a show_help message about it. N defaults to 25. N can be set to 0 disable the message via btl_usnic_max_short_packets. 1. If we receive a completion error for something other than a receive, display a show_help message. Reviewed by Dave Goodell. CMR'ing to v1.8.3, but it will require a custom patch because of the OMPI->OPAL BTL move. cmr=v1.8.3 This commit was SVN r32522.
2014-08-13 19:01:20 +04:00
/** how many short packets have to be received before outputting
the "received short packets" warning? */
uint32_t max_short_packets;
usnic: add btl_usnic_connectivity_map MCA param to output link information If the btl_usnic_connectivity_map MCA param is set to a non-NULL value, then each MPI process will output a file named <prefix>-<hostname>.pid<pid>.job<jobid>.mcwrank<MCW rank>.txt. Its contents will detail which usNIC device(s) (and therefore which link(s)) are being used to communicate with each peer MPI process. Here is a sample output file (named mpi005.pid26071.job1640759297.mcwrank0.txt): {{{ device=usnic_0,interface=eth4,ip=10.10.0.5/16,mac=24:57:20:05:20:00,mtu=9000 device=usnic_1,interface=eth5,ip=10.2.0.5/16,mac=24:57:20:05:21:00,mtu=9000 device=usnic_2,interface=eth6,ip=10.3.0.5/16,mac=24:57:20:05:50:00,mtu=9000 peer=1,hostname=mpi006,device=usnic_0@peer_ip=10.10.0.6/16@peer_mac=24:57:20:06:20:00,device=usnic_1@peer_ip=10.2.0.6/16@peer_mac=24:57:20:06:21:00,device=usnic_2@peer_ip=10.3.0.6/16@peer_mac=24:57:20:06:50:00 peer=2,hostname=mpi007,device=usnic_0@peer_ip=10.10.0.7/16@peer_mac=24:57:20:07:20:00,device=usnic_1@peer_ip=10.2.0.7/16@peer_mac=24:57:20:07:21:00,device=usnic_2@peer_ip=10.3.0.7/16@peer_mac=24:57:20:07:50:00 peer=3,hostname=mpi008,device=usnic_0@peer_ip=10.10.0.8/16@peer_mac=24:57:20:08:20:00,device=usnic_1@peer_ip=10.2.0.8/16@peer_mac=24:57:20:08:21:00,device=usnic_2@peer_ip=10.3.0.8/16@peer_mac=24:57:20:08:50:00 }}} Reviewed by Reese Faucette cmr=v1.8.2 This commit was SVN r32156.
2014-07-08 23:14:46 +04:00
/* Prefix for the connectivity map filename (map will be output if
the prefix is non-NULL) */
char *connectivity_map_prefix;
usnic: correctly count CRC errors Handle the differences between libfabric v1.0.0 and v1.1.0 in the return value of fi_cq_readerr(). Also consolidate CRC and truncation errors into the same handling block, since truncation errors are typically another symptom of CRC errors. This ensures that buffers get reposted properly.
2015-07-03 03:03:32 +03:00
/** Expected return value from fi_cq_readerr() upon success. In
libfabric v1.0.0 / API v1.0, the usnic provider returned
sizeof(fi_cq_err_entry) upon success. In libfabric >=v1.1 /
API >=v1.1, the usnic provider returned 1 upon success. */
ssize_t cq_readerr_success_value;
ssize_t cq_readerr_try_again_value;
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) 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.
2014-07-26 04:47:28 +04:00
} opal_btl_usnic_component_t;
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) 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.
2014-07-26 04:47:28 +04:00
OPAL_MODULE_DECLSPEC extern opal_btl_usnic_component_t mca_btl_usnic_component;
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) 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.
2014-07-26 04:47:28 +04:00
typedef mca_btl_base_recv_reg_t opal_btl_usnic_recv_reg_t;
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
/**
* Size for sequence numbers (just to ensure we use the same size
* everywhere)
*/
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) 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.
2014-07-26 04:47:28 +04:00
typedef uint16_t opal_btl_usnic_seq_t;
usnic: Shrink sequence numbers to 16 bits Authored-by: Reese Faucette <rfaucett@cisco.com> Reviewed-by: Jeff Squyres <jsquyres@cisco.com> cmr=v1.7.5:ticket=trac:4253 This commit was SVN r30834. The following Trac tickets were found above: Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
2014-02-26 11:40:10 +04:00
#define UDSEQ PRIu16
/* sequence number comparison macros that allow for rollover.
* Relies on the fact that sequence numbers should be relatively close
* together as compared to (1<<31)
*/
#define SEQ_DIFF(A,B) ((int16_t)((A)-(B)))
#define SEQ_LT(A,B) (SEQ_DIFF(A,B) < 0)
#define SEQ_LE(A,B) (SEQ_DIFF(A,B) <= 0)
#define SEQ_GT(A,B) (SEQ_DIFF(A,B) > 0)
#define SEQ_GE(A,B) (SEQ_DIFF(A,B) >= 0)
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
/**
* Register the usnic BTL MCA params
*/
usnic: remove all trailing whitespace Style cleanup only; no code changes. This commit was SVN r32366.
2014-07-31 00:56:15 +04:00
int opal_btl_usnic_component_register(void);
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
usnic: add ompi_btl_usnic_component_debug helper This new routine can be called in exceptional situations, either conditionally in BTL code or from a debugger, to help with debugging in cases where MSGDEBUG1/2 or stats logging are impractical but more detail is needed. Reviewed-by: Jeff Squyres <jsquyres@cisco.com> This commit was SVN r29483.
2013-10-23 19:51:11 +04:00
/**
* Routine which can be called from a debugger to print module, endpoint,
* fragment, and segment state to standard output. */
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) 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.
2014-07-26 04:47:28 +04:00
void opal_btl_usnic_component_debug(void);
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
usnic: add btl_usnic_connectivity_map MCA param to output link information If the btl_usnic_connectivity_map MCA param is set to a non-NULL value, then each MPI process will output a file named <prefix>-<hostname>.pid<pid>.job<jobid>.mcwrank<MCW rank>.txt. Its contents will detail which usNIC device(s) (and therefore which link(s)) are being used to communicate with each peer MPI process. Here is a sample output file (named mpi005.pid26071.job1640759297.mcwrank0.txt): {{{ device=usnic_0,interface=eth4,ip=10.10.0.5/16,mac=24:57:20:05:20:00,mtu=9000 device=usnic_1,interface=eth5,ip=10.2.0.5/16,mac=24:57:20:05:21:00,mtu=9000 device=usnic_2,interface=eth6,ip=10.3.0.5/16,mac=24:57:20:05:50:00,mtu=9000 peer=1,hostname=mpi006,device=usnic_0@peer_ip=10.10.0.6/16@peer_mac=24:57:20:06:20:00,device=usnic_1@peer_ip=10.2.0.6/16@peer_mac=24:57:20:06:21:00,device=usnic_2@peer_ip=10.3.0.6/16@peer_mac=24:57:20:06:50:00 peer=2,hostname=mpi007,device=usnic_0@peer_ip=10.10.0.7/16@peer_mac=24:57:20:07:20:00,device=usnic_1@peer_ip=10.2.0.7/16@peer_mac=24:57:20:07:21:00,device=usnic_2@peer_ip=10.3.0.7/16@peer_mac=24:57:20:07:50:00 peer=3,hostname=mpi008,device=usnic_0@peer_ip=10.10.0.8/16@peer_mac=24:57:20:08:20:00,device=usnic_1@peer_ip=10.2.0.8/16@peer_mac=24:57:20:08:21:00,device=usnic_2@peer_ip=10.3.0.8/16@peer_mac=24:57:20:08:50:00 }}} Reviewed by Reese Faucette cmr=v1.8.2 This commit was SVN r32156.
2014-07-08 23:14:46 +04:00
/**
* Called to output the connectivity map
*/
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) 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.
2014-07-26 04:47:28 +04:00
void opal_btl_usnic_connectivity_map(void);
usnic: add btl_usnic_connectivity_map MCA param to output link information If the btl_usnic_connectivity_map MCA param is set to a non-NULL value, then each MPI process will output a file named <prefix>-<hostname>.pid<pid>.job<jobid>.mcwrank<MCW rank>.txt. Its contents will detail which usNIC device(s) (and therefore which link(s)) are being used to communicate with each peer MPI process. Here is a sample output file (named mpi005.pid26071.job1640759297.mcwrank0.txt): {{{ device=usnic_0,interface=eth4,ip=10.10.0.5/16,mac=24:57:20:05:20:00,mtu=9000 device=usnic_1,interface=eth5,ip=10.2.0.5/16,mac=24:57:20:05:21:00,mtu=9000 device=usnic_2,interface=eth6,ip=10.3.0.5/16,mac=24:57:20:05:50:00,mtu=9000 peer=1,hostname=mpi006,device=usnic_0@peer_ip=10.10.0.6/16@peer_mac=24:57:20:06:20:00,device=usnic_1@peer_ip=10.2.0.6/16@peer_mac=24:57:20:06:21:00,device=usnic_2@peer_ip=10.3.0.6/16@peer_mac=24:57:20:06:50:00 peer=2,hostname=mpi007,device=usnic_0@peer_ip=10.10.0.7/16@peer_mac=24:57:20:07:20:00,device=usnic_1@peer_ip=10.2.0.7/16@peer_mac=24:57:20:07:21:00,device=usnic_2@peer_ip=10.3.0.7/16@peer_mac=24:57:20:07:50:00 peer=3,hostname=mpi008,device=usnic_0@peer_ip=10.10.0.8/16@peer_mac=24:57:20:08:20:00,device=usnic_1@peer_ip=10.2.0.8/16@peer_mac=24:57:20:08:21:00,device=usnic_2@peer_ip=10.3.0.8/16@peer_mac=24:57:20:08:50:00 }}} Reviewed by Reese Faucette cmr=v1.8.2 This commit was SVN r32156.
2014-07-08 23:14:46 +04:00
First commit of the Cisco usNIC BTL. This BTL accesses the Cisco usNIC Linux device via the Linux verbs API via Unreliable Datagram queue pairs. A few noteworthy points: * This BTL does most of its own fragmentation; it tells the PML that it has a very high max_send_size (much higher than the network MTU). * Since UD fragments are, by definition, unreliable, the usnic BTL handles all of its own reliability via a sliding window approach using the opal_hotel construct and many tricks stolen from the corpus of knowledge surrounding efficient TCP. * There is a fun PML latency-metric based optimization for NUMA awareness of short messages. * Note that this is ''not'' a generic UD verbs BTL; it is specific to the Cisco usNIC device. This commit was SVN r28879.
2013-07-20 02:13:58 +04:00
END_C_DECLS
#endif
Копировать постоянную ссылку