Use faster cl_ptr_vector_get() call instead of cl_ptr_vector_at(). In
this way eliminate 'stat' variable needs.
Signed-off-by: Sasha Khapyorsky sas...@voltaire.com
---
opensm/opensm/osm_subnet.c |8
1 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/opensm/opensm
Speedup a port lookup over LMC array - it is not necessary to match LMC
exactly for found port because base lid should be equal to requested lid
masked value, so '=' comparison should be enough and we don't need to
loop up to an actual port's lmc value match.
Signed-off-by: Sasha Khapyorsky sas
Since subn-port_lid_tbl vector is filled for all port's LIDs in
accordance with its LMC value, so we don't need to bother with LMC
tracking and instead can just return a pointer indexed by requested lid.
Obviously it speeds this helper up significantly.
Signed-off-by: Sasha Khapyorsky sas
On 14:45 Wed 23 Sep , Ira Weiny wrote:
From: Ira Weiny wei...@llnl.gov
Date: Wed, 23 Sep 2009 14:26:55 -0700
Subject: [PATCH] infiniband-diags/src/ibqueryerrors.c: fix bug when
attempting a sub-fabric scan
Also ibd_sm_id is never valid in this tool as the -s option is used
On 14:45 Wed 23 Sep , Ira Weiny wrote:
From: Ira Weiny wei...@llnl.gov
Date: Wed, 23 Sep 2009 14:26:55 -0700
Subject: [PATCH] infiniband-diags/src/ibqueryerrors.c: fix bug when
attempting a sub-fabric scan
Also ibd_sm_id is never valid in this tool as the -s option is used
Hi Ira,
On 15:09 Wed 23 Sep , Ira Weiny wrote:
From: Ira Weiny wei...@llnl.gov
Date: Wed, 23 Sep 2009 11:38:11 -0700
Subject: [PATCH] infiniband-diags/src/ibqueryerrors.c: Remove --all option
and replace it with --switch, --ca, --router
By default ibqueryerrors should print
Siplify node_type_to_print setup.
Signed-off-by: Sasha Khapyorsky sas...@voltaire.com
---
infiniband-diags/src/ibqueryerrors.c | 13 +
1 files changed, 5 insertions(+), 8 deletions(-)
diff --git a/infiniband-diags/src/ibqueryerrors.c
b/infiniband-diags/src/ibqueryerrors.c
index
On 12:06 Tue 29 Sep , Jeff Becker wrote:
Hi all. I propose the following plan to shutdown the general list:
1) unsubscribe all current subscribers
2) set the list to discard any incoming messages with an auto-discard
message that points you to linux-r...@vger.kernel.org
Seems as a good
Hi Hal,
On 15:34 Thu 24 Sep , Hal Rosenstock wrote:
When multiple switches are unlinked and then a switch is relinked,
it should behave like a cable pull or power down of switch so it
depends on the state of the remote peer port (as to linked or not).
This is not represented in the IB
On 16:20 Fri 25 Sep , Jim Schutt wrote:
On Fri, 2009-09-25 at 07:52 -0600, Sasha Khapyorsky wrote:
Hi Jim,
On 14:29 Tue 22 Sep , Jim Schutt wrote:
I'm working on another routing engine
That is interesting. What will be a key features of this new routing
engine
On 14:38 Tue 22 Sep , Hal Rosenstock wrote:
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
---
diff --git a/opensm/opensm/osm_mesh.c b/opensm/opensm/osm_mesh.c
index 260e2f8..beb6bd7 100644
--- a/opensm/opensm/osm_mesh.c
+++ b/opensm/opensm/osm_mesh.c
@@ -1565,6 +1565,39 @@
On 13:20 Thu 17 Sep , Ira Weiny wrote:
Sasha, would you be willing to accept such a patch? First move ib_types.h to
umad and then move the long inline functions into the lib and separate out
the remaining header.
Or would you prefer a new library? I think there is enough code there
Hi Jim,
On 14:29 Tue 22 Sep , Jim Schutt wrote:
I'm working on another routing engine
That is interesting. What will be a key features of this new routing
engine?
that also uses osm_switch_t:priv
to point to data that persists between calls to the routing engine, as
LASH does. And
When LASH runs its switch structures cleanup OpenSM can rediscover a
subnet and 'p_sw' pointer may refer already freed memory, so don't touch
it, just free our own stuff. (Note also that for valids OpenSM switches
objects' 'priv' pointers are cleared on lash_cleanup()).
Signed-off-by: Sasha
On 19:59 Tue 22 Sep , Hal Rosenstock wrote:
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
Applied. Thanks.
Sasha
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To
On 07:00 Wed 23 Sep , Hal Rosenstock wrote:
Let me try this again... The port LID table is fine but the lookup is done
based on the LID in the received portInfo as it is the result of
osm_physp_get_base_lid() (osm_link_mgr.c:link_mgr_get_smsl line 83).
Ok, makes sense.
Assuming that
- we don't need to
refer unintialized and not used status value when error response is
generated after osm_mgrp_add_port() failure. IOW:
commit 4dd928b705024c4fefd6435c733ddd885fded5ab
Author: Sasha Khapyorsky sas...@voltaire.com
Date: Tue Sep 22 18:31:13 2009 +0300
opensm
On 09:46 Tue 22 Sep , Hal Rosenstock wrote:
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
Applied. Thanks.
Sasha
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To
On 09:41 Mon 21 Sep , Hal Rosenstock wrote:
The intent of this change is to remove edge nodes (by not counting
them).
The point of this heuristic is to deal with the case of small
lattices which can easily have more surface than interior,
which leads to choosing a non representative
Hi Jim,
On 13:08 Fri 28 Aug , Jim Schutt wrote:
When LASH is run against ibsim, valgrind reports the following
(on x86_64) after a switch is removed from the fabric:
==15699== Invalid write of size 8
==15699==at 0x45FD8A: switch_delete (osm_ucast_lash.c:648)
==15699==by
Hi Doron,
On 15:03 Mon 21 Sep , Doron Shoham wrote:
Add ibcheckroutes script.
Wouldn't it be better to implement this using C program with help of
newly introduced libibnetdisc library? Saving all subsequent ibtracert
calls should improve performance dramatically.
Other comments are below.
On 10:41 Mon 14 Sep , Hal Rosenstock wrote:
How are the leaf switches determined (from core switches) in the
ibnetdiscover output ?
Using this:
+# find all leaf switches LIDs
+LIDS=($(awk '/# lid /{a[$(NF-1)]=$(NF-1)} END{for(v in a) print v}'
$topofile))
It parses out remote switch
On 07:32 Thu 17 Sep , Hal Rosenstock wrote:
Is that (lids in place) always the case ?
I don't see immediately how it could be not.
What about if the sets of PortInfo
for LID fail.
Set can fail, but internal OpenSM port_lid_tbl will be up to date.
Sasha
On 16:44 Tue 22 Sep , Hal Rosenstock wrote:
Yeah, the port lid table will be OK but port's PortInfo won't (so base
LID/LMC will be broken) for this scenario but it wouldn't affect this code
in this way. So I don't have any theories as to how this could occur. Do you
?
I don't, but if
On 10:36 Wed 02 Sep , Hal Rosenstock wrote:
Per published MgtWG errata
RefID 4484 - vendor specific MADs table support
RefID 4626 - reverse path PKey support in PathRecord responses
RefID 4635 - multicast FDB top support
RefID 4644 - hierarchy support
Signed-off-by: Hal Rosenstock
On 10:31 Wed 02 Sep , Hal Rosenstock wrote:
Add support for SwitchInfo:MulticastFDBTop
Added by MgtWG errata #4505-4508
Also, per MgtWG RefID #4640, MulticastFDBTop value of 0xbfff means no entries
In osm_mcast_mgr.c:mcast_mgr_set_mftables call new routine
mcast_mgr_set_mfttop to set
On 09:20 Mon 21 Sep , Hal Rosenstock wrote:
snip...
+
+ /*
+ * Do a blocking query for the PathRecord.
+ */
+ status = osmtest_get_path_rec_by_lid_pair(p_osmt, slid, dlid,
context);
+ if (status != IB_SUCCESS) {
+ OSM_LOG(p_osmt-log,
On 08:51 Mon 21 Sep , Hal Rosenstock wrote:
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
---
Changes since v1:
Fix use after free issue
diff --git a/opensm/opensm/osm_perfmgr_db.c b/opensm/opensm/osm_perfmgr_db.c
index e5dfc19..03f988d 100644
---
On 09:18 Mon 21 Sep , Hal Rosenstock wrote:
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
Applied. Thanks.
Sasha
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To
On 10:32 Mon 21 Sep , Hal Rosenstock wrote:
+ goto Exit;
+ }
It is not really stress testing, just pinging.
So are the other tests (additionally those use RMPP). Isn't repetitive
pinging a stress of a kind ?
No. Stress test assumes full load.
On 10:44 Mon 21 Sep , Hal Rosenstock wrote:
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
Applied. Thanks.
Sasha
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To
On 10:20 Mon 21 Sep , Hal Rosenstock wrote:
+
+ if (p_physp-port_info.capability_mask
IB_PORT_CAP_HAS_MCAST_FDB_TOP) {
BTW any reason why this capability bit if placed in PortInfo and not in
SwitchInfo (it is not port but switch related feature)?
I don't recall.
Could
On 11:05 Mon 21 Sep , Hal Rosenstock wrote:
On Mon, Sep 21, 2009 at 10:57 AM, Sasha Khapyorsky sas...@voltaire.comwrote:
snip...
How this ping test's timeline looks?
(1) client sends one request, (2) it travels to a server, (3) server
processes it and replies, (4) the response
On 11:22 Mon 21 Sep , Hal Rosenstock wrote:
I'll try.
Thanks.
I removed this in the latest patch version.
Ok. Let's wait with this patch up to capability mask
clarification/resolution.
Sasha
___
general mailing list
On 12:50 Mon 21 Sep , Hal Rosenstock wrote:
Other than this, is the patch acceptable ? I want to get this as ready as
possible.
I will look at v2.
Sasha
___
general mailing list
general@lists.openfabrics.org
On 10:38 Mon 21 Sep , Hal Rosenstock wrote:
Add support for SwitchInfo:MulticastFDBTop
Added by MgtWG errata #4505-4508
Also, per MgtWG RefID #4640, MulticastFDBTop value of 0xbfff means no entries
In osm_sm.c:osm_sm_set_mcast_tbl, when switch port 0 indicates
IsMulticastFDBTop
Hi Hal,
On 15:21 Mon 31 Aug , Hal Rosenstock wrote:
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
---
diff --git a/opensm/man/osmtest.8 b/opensm/man/osmtest.8
index fa0cd52..f0d6323 100644
--- a/opensm/man/osmtest.8
+++ b/opensm/man/osmtest.8
@@ -1,4 +1,4 @@
-.TH OSMTEST 8
On 08:55 Tue 01 Sep , Hal Rosenstock wrote:
Per published MgtWG errata:
RefID 4484 - vendor specific MADs
RefID 4575 - multicast PKey trap suppression
RefID 4641 - hierarchy info
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
Applied. Thanks.
Sasha
Improve some function prototypes:
- osm_mgrp_new() will get MCMember record now.
- osm_mgrp_is_port_present() is replaced by cleaner
osm_mgrp_get_mcm_port() helper.
Signed-off-by: Sasha Khapyorsky sas...@voltaire.com
---
opensm/include/opensm/osm_multicast.h | 25
When PathRecord query has multicast destination and SLID and/or SGID is
specified we need to ensure that this path source port is member of the
destination multicast group.
Signed-off-by: Sasha Khapyorsky sas...@voltaire.com
---
opensm/opensm/osm_sa_path_record.c | 14 ++
1 files
IBA 1.2.1 states (Vol.1, 15.2.5.16, p.918) that DGID shell be explicitly
specified when path destination of SA PathRecord query is multicast
group.
Signed-off-by: Sasha Khapyorsky sas...@voltaire.com
---
opensm/opensm/osm_sa_path_record.c | 45
1 files
This new helper is similar to osm_get_port_by_base_lid(), but with
improved interface (returns port's pointer) and moved to osm_subnet.[ch]
(where others osm_get_*_by_*() helpers are located).
Signed-off-by: Sasha Khapyorsky sas...@voltaire.com
---
opensm/include/opensm/osm_port.h | 35
On 09:32 Fri 18 Sep , Hal Rosenstock wrote:
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
Applied. Thanks.
Sasha
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To
Hi Hal,
On 10:31 Fri 18 Sep , Hal Rosenstock wrote:
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
---
diff --git a/opensm/opensm/osm_perfmgr_db.c b/opensm/opensm/osm_perfmgr_db.c
index e5dfc19..329743a 100644
--- a/opensm/opensm/osm_perfmgr_db.c
+++
On 17:09 Tue 15 Sep , Keshetti Mahesh wrote:
Cosmetic changes
Signed-off-by: Keshetti Mahesh keshetti.mah...@gmail.com
The patch is malformed (broken long lines and mangled whitespaces).
Applied by hands. Thanks.
Sasha
---
diff --git a/infiniband-diags/scripts/ibcheckwidth.in
On 09:12 Wed 16 Sep , Keshetti Mahesh wrote:
Fix: Modify ibcheckportwidth to exit if LWS is 1X instead of processing
next lines.
Trivial: ibcheckportwidth man page cosmetic change
Again, the patch is malformed. Please verify the patches before
submission.
Signed-off-by: Keshetti Mahesh
On 11:42 Mon 14 Sep , Hal Rosenstock wrote:
Also, similar change to doc/partition-config.txt
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
Applied. Thanks.
Sasha
___
general mailing list
general@lists.openfabrics.org
On 18:11 Tue 08 Sep , Hal Rosenstock wrote:
Also, update man page
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
Applied. Thanks.
Sasha
___
general mailing list
general@lists.openfabrics.org
On 09:04 Wed 09 Sep , Hal Rosenstock wrote:
Issuer GID is uninteresting for SM generated notices
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
Applied. Thanks.
Sasha
___
general mailing list
general@lists.openfabrics.org
Move report notice logging to separate function log_notice().
Signed-off-by: Sasha Khapyorsky sas...@voltaire.com
---
opensm/opensm/osm_inform.c | 70 +++
1 files changed, 37 insertions(+), 33 deletions(-)
diff --git a/opensm/opensm/osm_inform.c b
.
Signed-off-by: Doron Shoham dor...@voltaire.com
Signed-off-by: Sasha Khapyorsky sas...@voltaire.com
diff --git a/infiniband-diags/man/ibportstate.8
b/infiniband-diags/man/ibportstate.8
index b64c18d..fea860e 100644
--- a/infiniband-diags/man/ibportstate.8
+++ b/infiniband-diags/man/ibportstate.8
On 12:59 Mon 14 Sep , sebastien dugue wrote:
When grouping is enabled add a '#' separator between the switch guid
and the chassis type and slot.
I'm fine with this change, but please next time explain the change reason
too - this modifies ibnetdiscover output and may break
(reproducible under stress testing).
This patch fixes this by moving multicast group cleanup call under some
lock protected block as port removing code.
Signed-off-by: Sasha Khapyorsky sas...@voltaire.com
---
opensm/opensm/osm_sa_mcmember_record.c |4 +---
1 files changed, 1 insertions(+), 3 deletions
) multicast routing (MFTs) recalculation.
This also fixes a bug already reported (by me) on the list when due to
race between different SA requests processors during MCMember leave
storming multicast group cleanup code can double free an objects and
crash.
Signed-off-by: Sasha Khapyorsky sas
On 16:44 Fri 18 Sep , Sasha Khapyorsky wrote:
This patch fixes this by moving multicast group cleanup call under some
lock protected block as port removing code.
Please disregard this patch - it is not against mainstream where problem
exists, but not so trivially fixable. I will fix
bug (reported by Eli
couple of months ago) automatically:
http://lists.openfabrics.org/pipermail/general/2009-May/059644.html
Signed-off-by: Sasha Khapyorsky sas...@voltaire.com
---
opensm/include/opensm/osm_mcm_port.h | 18 ++--
opensm/include/opensm/osm_port.h | 81
Fix uninitialized return value in osm_sm_mcgrp_leave() function.
Signed-off-by: Sasha Khapyorsky sas...@voltaire.com
---
opensm/opensm/osm_sm.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/opensm/opensm/osm_sm.c b/opensm/opensm/osm_sm.c
index e446c9d..f3fa7f4 100644
On 08:45 Mon 14 Sep , Hal Rosenstock wrote:
Does this mean consolidate_ipv6_snm_req does not work now ?
No, it doesn't. As you may remember 'consolidate_ipv6_snm_req'
workaround does nothing with MGIDs to MLID mapping, but instead
enforces all IPv6 SNM matching requests to join a single
This adds port object reference to port related structures list in
multicast group. In this way we are saving couple of lookups over port
guid table and simplifying some interfaces.
Signed-off-by: Sasha Khapyorsky sas...@voltaire.com
---
Changes from v1:
Unify osm_mgrp_remove_port
On 07:34 Fri 11 Sep , Hal Rosenstock wrote:
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
Applied. Thanks.
Sasha
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To
Hi Hal,
On 18:23 Tue 08 Sep , Hal Rosenstock wrote:
On Sun, Sep 6, 2009 at 1:39 PM, Sasha Khapyorsky sas...@voltaire.comwrote:
When we have two or more changes in a same multicast group multiple
multicast rerouting requests will be created and processed. To prevent
this we will use
On 21:27 Wed 09 Sep , Eitan Zahavi wrote:
Sorry I am not following the exact details of the changes made.
Yes. The introduced patches don't reduce multicast routing scalability.
Sasha
___
general mailing list
general@lists.openfabrics.org
On 09:17 Thu 03 Sep , Hal Rosenstock wrote:
Also it says the default partition will be created “unconditionally even
when partition configuration file does not exist or cannot be accessed.”
Will it also be created if the partition configuration file exists but does
not have a default
On 16:43 Thu 03 Sep , Yevgeny Kliteynik wrote:
I've never done it this way but it does look like the partition create
code will detect the duplicated partitions (0x111 and 0x112) and merge
ports from rack2 with rack1 and rack4 with rack3.
It will.
Note that partition names are
This removes multicast group delayed deletion stuff - we don't need it
anymore because MFTs cleanup (in case of group removing) is performed
using MLID only (not mgrp content) which is requested by appropriate
osm_sm_mgrp_leave() call.
Signed-off-by: Sasha Khapyorsky sas...@voltaire.com
This adds port object reference to port related structures list in
multicast group. In this way we are saving couple of lookups over port
guid table and simplifying some interfaces.
Signed-off-by: Sasha Khapyorsky sas...@voltaire.com
---
opensm/include/opensm/osm_mcm_port.h | 24
On 18:25 Sun 06 Sep , Sasha Khapyorsky wrote:
Subject: [PATCH 1/2] Storage organization for multicast groups
BTW, I pushed now series of various multicast improvements. As far as I
understand this should simplify your implementation dramatically.
Sasha
On 20:49 Wed 19 Aug , Nicolas Morey-Chaisemartin wrote:
Except the first one, patches in this series are trivial cleanups.
The first one remove the reverse_hop parameter which is not used anymore
thanks to the current_hop.
Nicolas Morey-Chaisemartin (3):
osm_ucast_ftree.c: Removed
pointer
to multicast group object (mgrp). Such model makes it possible to
have MGIDs to MLID compression.
Signed-off-by: Sasha Khapyorsky sas...@voltaire.com
---
opensm/include/opensm/osm_mcm_info.h | 13 +++--
opensm/include/opensm/osm_port.h | 13 +++--
opensm/opensm
Use mgrp pointer to multicast group instead of mlid as parameter for
osm_sm_mcgrp_join/leave() functions. This simplifies the current
implementation, makes those functions MLID independent and in this way
helps to implement MGID to MLID compression.
Signed-off-by: Sasha Khapyorsky sas
Remove not used osm_mgrp_apply_func() function and associated types,
variables, etc..
Signed-off-by: Sasha Khapyorsky sas...@voltaire.com
---
opensm/include/opensm/osm_multicast.h | 66 -
opensm/opensm/osm_multicast.c | 38 ---
2 files
On 09:00 Thu 03 Sep , Hal Rosenstock wrote:
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
Applied. Thanks.
Sasha
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To
('to_be_deleted' flag) and will simplify many multicast
related code flows.
Signed-off-by: Sasha Khapyorsky sas...@voltaire.com
---
opensm/include/opensm/osm_sm.h |4 +-
opensm/opensm/osm_mcast_mgr.c | 27 +
opensm/opensm/osm_sm.c | 49
Following previous patch 'opensm: improve multicast re-routing requests
processing' remove multicast change id (last_change_id, last_tree_id),
we don't need to track it anymore i- all requests are processed only
once per mlid.
Signed-off-by: Sasha Khapyorsky sas...@voltaire.com
---
opensm
This patch renames mcast_mgr_process_mgrp() - mcast_mgr_process_mlid()
function and this gets MLID (in host byte order) as parameter. This
simplifies caller flows and makes the code ready for MGID to MLID
compression.
Signed-off-by: Sasha Khapyorsky sas...@voltaire.com
---
opensm/opensm
On 09:39 Mon 31 Aug , Hal Rosenstock wrote:
Similar to previous patch to Parallelize (Stripe) LFT sets across switches.
Currently, MADs are pipelined to a single switch first which effectively
serializes these requests. This patch pipelines the MFT set MADs across
switches first (before
On 10:42 Tue 01 Sep , Hal Rosenstock wrote:
Per published MgtWG errata:
RefID 4626 - reverse path PKey support in PathRecord responses
RefID 4635 - multicast FDB top support
RefID 4644 - hierarchy support
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
Applied. Thanks.
Sasha
On 13:42 Mon 31 Aug , Hal Rosenstock wrote:
Wouldn't endlid be set to top for this case (since top endlid) ? It
ignores endlid and not top in this case.
Ok, it is clear for me now (sorry, it took a long time :)). Thanks for
the explanations.
Sasha
On 12:35 Sun 30 Aug , Hal Rosenstock wrote:
Doesn't the loop:
for (block = startblock; block = lastblock; block++)
terminates without any blocks read ? So it shows no entries.
Sorry, I still don't understand. Let's suppose that top = 0xbfff,
cap = 1024, startlid = 0xc000, endlid = 0xc030
On 12:32 Sun 30 Aug , Hal Rosenstock wrote:
Add support for SwitchInfo:MulticastFDBTop
Added by MgtWG errata #4505-4508 and 4640
If MulticastFDBTop set to other than 0, only fetch MulticastForwardingTable
blocks up through MulticastFDBTop rather than MulticastFDBCap
If
On 18:59 Sat 29 Aug , Hal Rosenstock wrote:
diff --git a/opensm/opensm/osm_ucast_mgr.c b/opensm/opensm/osm_ucast_mgr.c
index b7e3893..39d825c 100644
--- a/opensm/opensm/osm_ucast_mgr.c
+++ b/opensm/opensm/osm_ucast_mgr.c
@@ -1007,6 +1007,7 @@ int osm_ucast_mgr_process(IN
On 14:23 Thu 06 Aug , Hal Rosenstock wrote:
In lash_process, rename variable from return_status to status
Also, status is not really IB_SUCCESS or not (although that works)
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
Applied. Thanks.
Sasha
Hi Hal,
On 18:34 Thu 06 Aug , Hal Rosenstock wrote:
@@ -773,6 +838,7 @@ static void seed_axes(lash_t *p_lash, int sw)
mesh_node_t *node = p_lash-switches[sw]-node;
int n = node-num_links;
int i, j, c;
+ char buf[256], *p;
OSM_LOG_ENTER(p_log);
if
Hi Hal,
On 12:41 Fri 07 Aug , Hal Rosenstock wrote:
Similar to previous patch to Parallelize (Stripe) LFT sets across switches.
Currently, MADs are pipelined to a single switch first which effectively
serializes these requests. This patch pipelines the MFT set MADs across
switches first
On 10:02 Wed 26 Aug , Hal Rosenstock wrote:
Add support for SwitchInfo:MulticastFDBTop and
PortInfo:CapabilityMask.IsMulticastFDBTopSupported
Added by MgtWG errata #4505-4508
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
Applied. Thanks.
Sasha
On 10:03 Wed 26 Aug , Hal Rosenstock wrote:
Add support for SwitchInfo:MulticastFDBTop
Added by MgtWG errata #4505-4508 and #4640
If MulticastFDBTop is set to other than 0, only fetch MulticastForwardingTable
blocks up through MulticastFDBTop rather than MulticastFDBCap
If
On 07:32 Sun 30 Aug , Hal Rosenstock wrote:
osm_link_mgr.c:link_mgr_get_smsl has the following:
/* Find osm_port of the source = p_physp */
slid = osm_physp_get_base_lid(p_physp);
p_src_port =
cl_ptr_vector_get(sm-p_subn-port_lid_tbl,
On 10:04 Wed 26 Aug , Hal Rosenstock wrote:
@@ -5899,6 +5899,8 @@ typedef struct _ib_switch_info {
ib_net16_t lids_per_port;
ib_net16_t enforce_cap;
uint8_t flags;
+ uint8_t resvd;
+ ib_net16_t mcast_top;
} PACK_SUFFIX ib_switch_info_t;
#include
On 17:37 Wed 26 Aug , Eli Dorfman (Voltaire) wrote:
Subject: [PATCH] Fix IB network discovery from switch node.
Signed-off-by: Eli Dorfman e...@voltaire.com
Applied. Thanks.
Please next time add descriptive change log to your patches.
Sasha
On 11:54 Wed 26 Aug , Hal Rosenstock wrote:
Per MgtWG RefID #4527
Also, cosmetic commentary change
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
Applied. Thanks.
Next time could you add more descriptive change log to your patches -
RefID #4527 by itself doesn't say a lot
On 15:11 Sun 30 Aug , Sasha Khapyorsky wrote:
On 17:37 Wed 26 Aug , Eli Dorfman (Voltaire) wrote:
Subject: [PATCH] Fix IB network discovery from switch node.
Signed-off-by: Eli Dorfman e...@voltaire.com
Applied. Thanks.
BTW, was need to rebase the patch against master.
Sasha
On 12:12 Wed 26 Aug , Hal Rosenstock wrote:
Indicate extended v. (normal) port counters in output
Also, some cosmetic formatting changes and commentary typo fixed
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
Applied. Thanks.
Sasha
On 10:29 Wed 26 Aug , Ira Weiny wrote:
From: Ira Weiny wei...@llnl.gov
Date: Fri, 21 Aug 2009 15:01:00 -0700
Subject: [PATCH] infiniband-diags/libibnetdisc: add missing '\n' to error
message
Signed-off-by: Ira Weiny wei...@llnl.gov
Applied. Thanks.
Sasha
On 10:31 Wed 26 Aug , Ira Weiny wrote:
From: Ira Weiny wei...@llnl.gov
Date: Wed, 26 Aug 2009 09:25:00 -0700
Subject: [PATCH] libibnetdisc: add retract_dpath function
When using combined routing some switches do not handle Hop Count of 0
well. Detect when the drpath count
On 08:22 Sun 30 Aug , Hal Rosenstock wrote:
Why should be pad[1] here? In struct switch_info you are adding three
bytes (resvd - 1 and mcast_top - 2), no?
Good catch. It was due to an initial version which didn't have the 16 bit
MFTTop alignment. Do you want a v2 patch for this ?
-by: Sasha Khapyorsky sas...@voltaire.com
---
infiniband-diags/libibnetdisc/src/ibnetdisc.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/infiniband-diags/libibnetdisc/src/ibnetdisc.c
b/infiniband-diags/libibnetdisc/src/ibnetdisc.c
index 0f6fc55..97e369c 100644
On 08:51 Sun 30 Aug , Hal Rosenstock wrote:
Add support for SwitchInfo:MulticastFDBTop
Added by MgtWG errata #4505-4508
Add OpenSM infrastructure support to ib_types.h and osm_helper.c
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
Applied. Thanks.
Sasha
On 08:42 Sun 30 Aug , Hal Rosenstock wrote:
This is handled by the block loop inside of dump_multicast_tables.
Where? I don't see this. Should not it to show nothing (no entries)
when top = 0xbfff and dump_all is not set?
Sasha
___
general
On 09:25 Sun 30 Aug , Hal Rosenstock wrote:
Add support for SwitchInfo:MulticastFDBTop
Added by MgtWG errata #4505-4508 and 4640
If MulticastFDBTop set to other than 0, only fetch MulticastForwardingTable
blocks up through MulticastFDBTop rather than MulticastFDBCap
If
On 12:03 Fri 28 Aug , Hal Rosenstock wrote:
lash_core: ERR 4D02: Lane requirements (9) exceed available lanes (8) with
starting lane (0)
ucast_mgr_route: lash: cannot build fwd tables.
osm_ucast_mgr_process: minhop tables configured on all switches
ERR 331D: LFT of switch 0xguid is not
1 - 100 of 2402 matches
Mail list logo