Repost of the previous set of patches:
Hi,
We are current working on a Ftree topology where IO nodes are connected on
spine switches.
Using the cn_guid_file and root_guid_file works great.
It is possible to route the whole tree as a fat tree. All the CNs are connected
to the other CN and IO nodes.
However, we are missing some connectivity between IO nodes. This is the
expected behavior as the route between those IO nodes would have
to go down to go back up on another spine switch.
However, we need at least a bit of connectivity between those nodes. There won't be any
real traffic but just some "ping" for HA purposes.
Therefore, I have implemented two new options to openSM: io_guid_file and
max_reverse_hops.
The io_guid_file provides a list of all the IO guid (it may differs from the
list of non-CN nodes)
The max_reverse_hops gives the number of time IO nodes (described by
io_guid_file) are allowed to use a switch backward.
According to my tests this has absolutely no effects on regular routing and
manages to connect the io nodes together, if max_reverse_hops is big enough.
Regards
Nicolas Morey- Chaisemartin
____
Nicolas Morey-Chaisemartin (3):
opensm: Added io_guid_file and max_reverse_hops options
opensm/osm_ucast_ftree.c: Added possible reverse hops for Ftree
algorithm.
Added documentation for io_guid_file and max_reverse_hop feature
opensm/doc/current-routing.txt | 32 +++++
opensm/include/opensm/osm_subnet.h | 6 +
opensm/man/opensm.8.in | 27 ++++
opensm/opensm/main.c | 26 ++++-
opensm/opensm/osm_subnet.c | 12 ++
opensm/opensm/osm_ucast_ftree.c | 244 +++++++++++++++++++++++++++++-------
6 files changed, 303 insertions(+), 44 deletions(-)
_______________________________________________
general mailing list
[email protected]
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
