ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
1d7d7c201b
openmpi/opal/mca/btl/usnic/btl_usnic_send.h

320 строки
10 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
/*
usnic: ensure all messages are sent on the data channel Messages should go on the data channel, even if they're short. Only ACKs go on the priority channel.
2016-01-30 18:34:20 +03:00
* Copyright (c) 2013-2016 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
*
* $HEADER$
*/
#ifndef BTL_USNIC_SEND_H
#define BTL_USNIC_SEND_H
#include "btl_usnic.h"
#include "btl_usnic_frag.h"
#include "btl_usnic_ack.h"
#if MSGDEBUG1
#include "btl_usnic_util.h"
#endif
/*
usnic: remove all trailing whitespace Style cleanup only; no code changes. This commit was SVN r32366.
2014-07-31 00:56:15 +04:00
* Check if conditions are right, and if so, put endpoint on
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 endpoints that have sends to be done
*/
static inline 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
opal_btl_usnic_check_rts(
opal_btl_usnic_endpoint_t *endpoint)
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
{
/*
* If endpoint not already ready,
* and has packets to send,
* and it has send credits,
* and its retransmission window is open,
* make it ready
*/
if (!endpoint->endpoint_ready_to_send &&
!opal_list_is_empty(&endpoint->endpoint_frag_send_queue) &&
endpoint->endpoint_send_credits > 0 &&
WINDOW_OPEN(endpoint)) {
opal_list_append(&endpoint->endpoint_module->endpoints_with_sends,
&endpoint->super);
endpoint->endpoint_ready_to_send = true;
#if MSGDEBUG1
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
opal_output(0, "make endpoint %p RTS\n", (void*)endpoint);
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 {
opal_output(0, "rts:%d empty:%d cred:%d open%d\n",
endpoint->endpoint_ready_to_send,
opal_list_is_empty(&endpoint->endpoint_frag_send_queue),
endpoint->endpoint_send_credits,
WINDOW_OPEN(endpoint));
#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
* Common point for posting a segment
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 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
opal_btl_usnic_post_segment(
opal_btl_usnic_module_t *module,
opal_btl_usnic_endpoint_t *endpoint,
opal_btl_usnic_send_segment_t *sseg)
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
{
int ret;
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
/* get channel and remote channel */
opal_btl_usnic_channel_id_t channel_id = sseg->ss_channel;
opal_btl_usnic_channel_t *channel = &module->mod_channels[channel_id];
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
#if MSGDEBUG1
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
opal_output(0, "post_send: type=%s, ep=%p, remote_addr=%p, addr=%p, len=%"
PRIsize_t,
usnic_seg_type_str(sseg->ss_base.us_type),
(void*) channel->ep,
(void*) endpoint->endpoint_remote_addrs[channel_id],
(void*) sseg->ss_ptr,
sseg->ss_len);
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
#endif
usnic: update to latest libfabric API changes
2014-12-04 01:44:50 +03:00
usnic: ensure all messages are sent on the data channel Messages should go on the data channel, even if they're short. Only ACKs go on the priority channel.
2016-01-30 18:34:20 +03:00
assert(channel_id == USNIC_DATA_CHANNEL);
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
/* Send the segment */
usnic: update to latest libfabric API changes
2014-12-04 01:44:50 +03:00
ret = fi_send(channel->ep,
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
sseg->ss_ptr,
usnic: handle FI_MSG_PREFIX differences libfabric v1.0.0->v1.1.0 In libfabric v1.0.0 (i.e., API v1.0), the usnic provider handled FI_MSG_PREFIX inconsistently between sends and receives. This has been fixed in libfabric v1.1.0 (i.e., API v1.1): FI_MSG_PREFIX is handled consistently for both sends and receives. Run-time detect which libfabric we are running with and adapt behavior appropriately.
2015-07-09 21:58:00 +03:00
sseg->ss_len + mca_btl_usnic_component.prefix_send_offset,
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
NULL,
endpoint->endpoint_remote_addrs[channel_id],
sseg);
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
if (OPAL_UNLIKELY(0 != ret)) {
usnic: update to latest libfabric API changes
2014-12-04 01:44:50 +03:00
opal_btl_usnic_util_abort("fi_send() failed",
usnic: do not call ompi_rte_abort() Adapt to moving down to OPAL: find a PML-registered error callback, and use that when we don't have a module context and we need to abort. Failing that, just call exit(). This commit was SVN r32362.
2014-07-31 00:52:06 +04:00
__FILE__, __LINE__);
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
/* Never returns */
}
/* track # of time non-ACKs are posted */
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
if (sseg->ss_base.us_type != OPAL_BTL_USNIC_SEG_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
++sseg->ss_send_posted;
++sseg->ss_parent_frag->sf_seg_post_cnt;
}
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
/* Stats */
++module->stats.num_total_sends;
++channel->num_channel_sends;
--channel->credits;
}
/*
* Common point for posting an ACK
*/
static inline void
opal_btl_usnic_post_ack(
opal_btl_usnic_module_t *module,
opal_btl_usnic_endpoint_t *endpoint,
opal_btl_usnic_send_segment_t *sseg)
{
int ret;
/* get channel and remote channel */
opal_btl_usnic_channel_id_t channel_id = sseg->ss_channel;
opal_btl_usnic_channel_t *channel = &module->mod_channels[channel_id];
#if MSGDEBUG1
opal_output(0, "post_send ACK: type=%s, ep=%p, remote_addr=%p, addr=%p, len=%"
PRIsize_t,
usnic_seg_type_str(sseg->ss_base.us_type),
(void*) channel->ep,
(void*) endpoint->endpoint_remote_addrs[channel_id],
(void*) sseg->ss_ptr,
sseg->ss_len);
#endif
usnic: ensure all messages are sent on the data channel Messages should go on the data channel, even if they're short. Only ACKs go on the priority channel.
2016-01-30 18:34:20 +03:00
assert(channel_id == USNIC_PRIORITY_CHANNEL);
usnic: update to latest libfabric API changes
2014-12-04 01:44:50 +03:00
ret = fi_send(channel->ep,
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
sseg->ss_ptr,
usnic: handle FI_MSG_PREFIX differences libfabric v1.0.0->v1.1.0 In libfabric v1.0.0 (i.e., API v1.0), the usnic provider handled FI_MSG_PREFIX inconsistently between sends and receives. This has been fixed in libfabric v1.1.0 (i.e., API v1.1): FI_MSG_PREFIX is handled consistently for both sends and receives. Run-time detect which libfabric we are running with and adapt behavior appropriately.
2015-07-09 21:58:00 +03:00
sseg->ss_len + mca_btl_usnic_component.prefix_send_offset,
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
NULL,
endpoint->endpoint_remote_addrs[channel_id],
sseg);
if (OPAL_UNLIKELY(0 != ret)) {
usnic: update to latest libfabric API changes
2014-12-04 01:44:50 +03:00
opal_btl_usnic_util_abort("fi_send() failed",
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
__FILE__, __LINE__);
/* Never returns */
}
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
/* Stats */
Move all usNIC stats to _stats.c|h and export them as MPI_T pvars. This commit moves all the module stats into their own struct so that the stats only need to appear as a single line in the module_t definition, and then moves all the logic for reporting the stats into btl_usnic_stats.c|h. Further, the stats are now exported as MPI_T_BIND_NO_OBJECT entities (i.e., not bound to any particular MPI handle), and are marked as READONLY and CONTINUOUS. They currently all default to verbose level 5 ("Application tuner / detailed", according to https://svn.open-mpi.org/trac/ompi/wiki/MCAParamLevels). Most of the statistics are counters, but a small number are high watermark values. Due to how counters are reported via MPI_T, none of the counters are exported through MPI_T if the MCA param btl_usnic_stats_relative=1 (i.e., the module resets the stats back to zero at a given frequency). When MPI_T_pvar_handle_alloc() is invoked on any of these pvars, it will return a count that is equal to the number of active usnic BTL modules. The values returned for any given pvar (e.g., num_total_sends) are an array containing one value for each active usnic BTL module. The ordering of values in the array is both consistent across all usnic pvars and stable throughout a single job: array slot 0 corresponds to module X, array slot 1 corresponds to module Y, etc. Mapping which array slot corresponds to which underlying Linux usnic_X device works as follows: * The btl_usnic_devices MPI_T state pvar is associated with a btl_usnic_device MPI_T enum, and be obtained via MPI_T_pvar_get_info(). * If all usNIC pvars are of length N, the values [0,N) in the btl_usnic_device enum are associated with strings of the corresponding underlying Linux device. For exampe, to look up which Linux device is reported in all usNIC pvars' array slot 1, look up the int value 1 in the btl_usnic_devices enum. Its corresponding string value is underlying Linux device name (e.g., "usnic_1"). cmr=v1.7.4:subject="usnic BTL MPI_T pvars" This commit was SVN r29545.
2013-10-29 02:23:08 +04:00
++module->stats.num_total_sends;
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
++channel->num_channel_sends;
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
--channel->credits;
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
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
* Post a send to the work queue
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 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
opal_btl_usnic_endpoint_send_segment(
opal_btl_usnic_module_t *module,
opal_btl_usnic_send_segment_t *sseg)
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_btl_usnic_send_frag_t *frag;
opal_btl_usnic_endpoint_t *endpoint;
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
uint16_t sfi;
frag = sseg->ss_parent_frag;
endpoint = frag->sf_endpoint;
/* Do we have room in the endpoint's sender window?
Sender window:
|-------- WINDOW_SIZE ----------|
+---------------------------------+
| next_seq_to_send |
| somewhere in this range |
^+---------------------------------+
|
+-- ack_seq_rcvd: one less than the window left edge
Assuming that next_seq_to_send is > ack_seq_rcvd (verified
by assert), then the good condition to send is:
next_seq_to_send <= ack_seq_rcvd + WINDOW_SIZE
And therefore the bad condition is
next_seq_to_send > ack_seq_rcvd + WINDOW_SIZE
*/
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
assert(SEQ_GT(endpoint->endpoint_next_seq_to_send,
endpoint->endpoint_ack_seq_rcvd));
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
assert(WINDOW_OPEN(endpoint));
/* Assign sequence number and increment */
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
sseg->ss_base.us_btl_header->pkt_seq =
endpoint->endpoint_next_seq_to_send++;
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
/* Fill in remote address to indicate PUT or not */
sseg->ss_base.us_btl_header->put_addr =
usnic: Fix compile errors related to r32196 This commit was SVN r32198. The following SVN revision numbers were found above: r32196 --> open-mpi/ompi@a14e0f10d4a083c66652f92a5300aef089058a08
2014-07-10 21:18:03 +04:00
frag->sf_base.uf_remote_seg[0].seg_addr.pval;
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
/* piggy-back an ACK if needed */
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_piggyback_ack(endpoint, sseg);
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
#if MSGDEBUG1
{
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
char local_ip[32];
char remote_ip[32];
opal_btl_usnic_snprintf_ipv4_addr(local_ip, sizeof(local_ip),
module->local_modex.ipv4_addr,
module->local_modex.netmask);
opal_btl_usnic_snprintf_ipv4_addr(remote_ip, sizeof(remote_ip),
endpoint->endpoint_remote_modex.ipv4_addr,
endpoint->endpoint_remote_modex.netmask);
opal_output(0, "--> Sending %s: seq: %" UDSEQ ", sender: 0x%016lx from device %s, IP %s, port %u, seg %p, room %d, wc len %u, remote IP %s, port %u",
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
(sseg->ss_parent_frag->sf_base.uf_type == OPAL_BTL_USNIC_FRAG_LARGE_SEND)?
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
"CHUNK" : "FRAG",
sseg->ss_base.us_btl_header->pkt_seq,
sseg->ss_base.us_btl_header->sender,
btl/usnic: update for libfabric v1.4 With libfabric v1.4, the usnic provider changed the values of its fabric and domain name strings (compared to libfabric <v1.4). Update the Open MPI usNIC BTL to handle both pre-v1.4 and v1.4 fabric/domain names. Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
2016-08-20 05:07:14 +03:00
endpoint->endpoint_module->linux_device_name,
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
local_ip,
module->local_modex.ports[sseg->ss_channel],
(void*)sseg,
sseg->ss_hotel_room,
sseg->ss_ptr,
remote_ip,
endpoint->endpoint_remote_modex.ports[sseg->ss_channel]);
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
}
#endif
/* do the actual send */
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_post_segment(module, endpoint, sseg);
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
/* Track this header by stashing in an array on the endpoint that
is the same length as the sender's window (i.e., WINDOW_SIZE).
To find a unique slot in this array, use (seq % WINDOW_SIZE).
*/
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
sfi = WINDOW_SIZE_MOD(sseg->ss_base.us_btl_header->pkt_seq);
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
endpoint->endpoint_sent_segs[sfi] = sseg;
sseg->ss_ack_pending = true;
/* bookkeeping */
--endpoint->endpoint_send_credits;
/* Stats */
if (sseg->ss_parent_frag->sf_base.uf_type
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_FRAG_LARGE_SEND) {
Move all usNIC stats to _stats.c|h and export them as MPI_T pvars. This commit moves all the module stats into their own struct so that the stats only need to appear as a single line in the module_t definition, and then moves all the logic for reporting the stats into btl_usnic_stats.c|h. Further, the stats are now exported as MPI_T_BIND_NO_OBJECT entities (i.e., not bound to any particular MPI handle), and are marked as READONLY and CONTINUOUS. They currently all default to verbose level 5 ("Application tuner / detailed", according to https://svn.open-mpi.org/trac/ompi/wiki/MCAParamLevels). Most of the statistics are counters, but a small number are high watermark values. Due to how counters are reported via MPI_T, none of the counters are exported through MPI_T if the MCA param btl_usnic_stats_relative=1 (i.e., the module resets the stats back to zero at a given frequency). When MPI_T_pvar_handle_alloc() is invoked on any of these pvars, it will return a count that is equal to the number of active usnic BTL modules. The values returned for any given pvar (e.g., num_total_sends) are an array containing one value for each active usnic BTL module. The ordering of values in the array is both consistent across all usnic pvars and stable throughout a single job: array slot 0 corresponds to module X, array slot 1 corresponds to module Y, etc. Mapping which array slot corresponds to which underlying Linux usnic_X device works as follows: * The btl_usnic_devices MPI_T state pvar is associated with a btl_usnic_device MPI_T enum, and be obtained via MPI_T_pvar_get_info(). * If all usNIC pvars are of length N, the values [0,N) in the btl_usnic_device enum are associated with strings of the corresponding underlying Linux device. For exampe, to look up which Linux device is reported in all usNIC pvars' array slot 1, look up the int value 1 in the btl_usnic_devices enum. Its corresponding string value is underlying Linux device name (e.g., "usnic_1"). cmr=v1.7.4:subject="usnic BTL MPI_T pvars" This commit was SVN r29545.
2013-10-29 02:23:08 +04:00
++module->stats.num_chunk_sends;
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 {
Move all usNIC stats to _stats.c|h and export them as MPI_T pvars. This commit moves all the module stats into their own struct so that the stats only need to appear as a single line in the module_t definition, and then moves all the logic for reporting the stats into btl_usnic_stats.c|h. Further, the stats are now exported as MPI_T_BIND_NO_OBJECT entities (i.e., not bound to any particular MPI handle), and are marked as READONLY and CONTINUOUS. They currently all default to verbose level 5 ("Application tuner / detailed", according to https://svn.open-mpi.org/trac/ompi/wiki/MCAParamLevels). Most of the statistics are counters, but a small number are high watermark values. Due to how counters are reported via MPI_T, none of the counters are exported through MPI_T if the MCA param btl_usnic_stats_relative=1 (i.e., the module resets the stats back to zero at a given frequency). When MPI_T_pvar_handle_alloc() is invoked on any of these pvars, it will return a count that is equal to the number of active usnic BTL modules. The values returned for any given pvar (e.g., num_total_sends) are an array containing one value for each active usnic BTL module. The ordering of values in the array is both consistent across all usnic pvars and stable throughout a single job: array slot 0 corresponds to module X, array slot 1 corresponds to module Y, etc. Mapping which array slot corresponds to which underlying Linux usnic_X device works as follows: * The btl_usnic_devices MPI_T state pvar is associated with a btl_usnic_device MPI_T enum, and be obtained via MPI_T_pvar_get_info(). * If all usNIC pvars are of length N, the values [0,N) in the btl_usnic_device enum are associated with strings of the corresponding underlying Linux device. For exampe, to look up which Linux device is reported in all usNIC pvars' array slot 1, look up the int value 1 in the btl_usnic_devices enum. Its corresponding string value is underlying Linux device name (e.g., "usnic_1"). cmr=v1.7.4:subject="usnic BTL MPI_T pvars" This commit was SVN r29545.
2013-10-29 02:23:08 +04:00
++module->stats.num_frag_sends;
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
}
/* If we have room in the sender's window, we also have room in
endpoint hotel */
opal_hotel_checkin_with_res(&endpoint->endpoint_hotel, sseg,
&sseg->ss_hotel_room);
}
/*
* This enqueues a fragment send into the system. A send of a fragment
* may result in the sending of multiple segments
*/
static inline int
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_endpoint_enqueue_frag(
opal_btl_usnic_endpoint_t *endpoint,
opal_btl_usnic_send_frag_t *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
{
#if MSGDEBUG1
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
opal_output(0, "enq_frag: frag=%p, endpoint=%p, %s, len=%lu\n",
Cisco CSCuj12520: various problems running c_fence_put_1 - tag needs to be sent in *our* header, not the PML header - usnic_alloc() should return smaller value if too much data requested - be careful about callbacks vs removing items from lists (we need to remove from outr lists *before* the callback) - improve send callback handling - add some more MSGDEBUG2 logging and cleanup This commit was SVN r29181.
2013-09-17 11:20:44 +04:00
(void*)frag, (void*)endpoint,
usnic_frag_type(frag->sf_base.uf_type),
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
(long unsigned)frag->sf_base.uf_base.des_src->seg_len);
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
if (frag->sf_base.uf_type == OPAL_BTL_USNIC_FRAG_LARGE_SEND) {
opal_btl_usnic_large_send_frag_t *lfrag;
lfrag = (opal_btl_usnic_large_send_frag_t *)frag;
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
opal_output(0, " large size=%zd\n", lfrag->lsf_base.sf_size);
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
}
#endif
/* add to tail of in-progress list */
opal_list_append(&endpoint->endpoint_frag_send_queue,
&frag->sf_base.uf_base.super.super);
/* possibly make this endpoint ready to send again */
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_check_rts(endpoint);
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
return OPAL_SUCCESS;
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
}
Cisco CSCuj22803: fixes for Bsend changes required to support MPI_Bsend(). Introduces concept of attaching a buffer to a large segment that the PML can scribble into and we will send from. The reason we don't use a pinned buffer and send directly from that is that usnic_verbs does not (yes) support num_sge>1 for regular sends. This means the data gets copied twice, but that is unavoidable. changed the logic in handle_large_send to be more sensible Incorporated David's review comments This commit was SVN r29184.
2013-09-17 11:27:39 +04:00
static inline 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
opal_btl_usnic_release_send_segment(
opal_btl_usnic_module_t *module,
opal_btl_usnic_send_frag_t *frag,
opal_btl_usnic_send_segment_t *sseg)
Cisco CSCuj22803: fixes for Bsend changes required to support MPI_Bsend(). Introduces concept of attaching a buffer to a large segment that the PML can scribble into and we will send from. The reason we don't use a pinned buffer and send directly from that is that usnic_verbs does not (yes) support num_sge>1 for regular sends. This means the data gets copied twice, but that is unavoidable. changed the logic in handle_large_send to be more sensible Incorporated David's review comments This commit was SVN r29184.
2013-09-17 11:27:39 +04:00
{
usnic: remove all trailing whitespace Style cleanup only; no code changes. This commit was SVN r32366.
2014-07-31 00:56:15 +04:00
/* We only return CHUNK segments because they are the only send-style
Cisco CSCuj22803: fixes for Bsend changes required to support MPI_Bsend(). Introduces concept of attaching a buffer to a large segment that the PML can scribble into and we will send from. The reason we don't use a pinned buffer and send directly from that is that usnic_verbs does not (yes) support num_sge>1 for regular sends. This means the data gets copied twice, but that is unavoidable. changed the logic in handle_large_send to be more sensible Incorporated David's review comments This commit was SVN r29184.
2013-09-17 11:27:39 +04:00
* segments that are dynamically allocated.
*/
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
if (sseg->ss_base.us_type == OPAL_BTL_USNIC_SEG_CHUNK) {
opal_btl_usnic_chunk_segment_return(module, sseg);
Cisco CSCuj22803: fixes for Bsend changes required to support MPI_Bsend(). Introduces concept of attaching a buffer to a large segment that the PML can scribble into and we will send from. The reason we don't use a pinned buffer and send directly from that is that usnic_verbs does not (yes) support num_sge>1 for regular sends. This means the data gets copied twice, but that is unavoidable. changed the logic in handle_large_send to be more sensible Incorporated David's review comments This commit was SVN r29184.
2013-09-17 11:27:39 +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
void opal_btl_usnic_frag_complete(opal_btl_usnic_send_frag_t *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
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_frag_send_complete(opal_btl_usnic_module_t *module,
opal_btl_usnic_send_segment_t *sseg);
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
void opal_btl_usnic_chunk_send_complete(opal_btl_usnic_module_t *module,
opal_btl_usnic_send_segment_t *sseg);
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: pack via convertor on the fly If we need to use a convertor, go back to stashing that convertor in the frag and populating segments "on the fly" (in ompi_btl_usnic_module_progress_sends). Previously we would pack into a chain of chunk segments at prepare_src time, unnecessarily consuming additional memory. Reviewed-by: Jeff Squyres <jsquyres@cisco.com> Reviewed-by: Reese Faucette <rfaucett@cisco.com> This commit was SVN r29592.
2013-11-05 02:52:03 +04:00
int
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_finish_put_or_send(
opal_btl_usnic_module_t *module,
opal_btl_usnic_endpoint_t *endpoint,
opal_btl_usnic_send_frag_t *frag,
usnic: pack via convertor on the fly If we need to use a convertor, go back to stashing that convertor in the frag and populating segments "on the fly" (in ompi_btl_usnic_module_progress_sends). Previously we would pack into a chain of chunk segments at prepare_src time, unnecessarily consuming additional memory. Reviewed-by: Jeff Squyres <jsquyres@cisco.com> Reviewed-by: Reese Faucette <rfaucett@cisco.com> This commit was SVN r29592.
2013-11-05 02:52:03 +04:00
mca_btl_base_tag_t tag)
__opal_attribute_noinline__;
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
#endif /* BTL_USNIC_SEND_H */
Копировать постоянную ссылку