p_line allocated by getline() isn't being freed at the end.
Signed-off-by: Alex Netes ale...@voltaire.com
---
opensm/opensm/osm_console_io.c |2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/opensm/opensm/osm_console_io.c b/opensm/opensm/osm_console_io.c
index 0a15313
On 15:19 Fri 11 Feb , Mike Heinz wrote:
Alex,
Given the change over in libibumad, you may not have seen this, so I'm
resending it.
-Original Message-
From: Mike Heinz
Sent: Monday, February 07, 2011 11:10 AM
To: linux-rdma@vger.kernel.org; ewg
Cc: 'Sasha Khapyorsky'
We finished the management tree seperation.
From now on, Ira Weiny wei...@llnl.gov takes the responsibility for
maitaining libibmad and
infiniband-diags. His trees are:
git://git.openfabrics.org/~iraweiny/libibmad
git://git.openfabrics.org/~iraweiny/infiniband-diags
libibumad, opensm
On 12:53 Thu 10 Feb , Jim Schutt wrote:
From: Hal Rosenstock h...@mellanox.com
Signed-off-by: Hal Rosenstock h...@mellanox.com
Signed-off-by: Jim Schutt jasc...@sandia.gov
Applied. Thanks.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a message to
On 21:24 Tue 28 Dec , Hal Rosenstock wrote:
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 supported, set MulticastFDBTop in SwitchInfo
based
c2755aa360d3f29d04865ba4e2454a98 libibmad-1.3.7.tar.gz
c7575b7620615d7dfa1c7fdbbd310ec7 libibumad-1.3.7.tar.gz
df051f5f0192d369b0b904147cb045a8 opensm-3.3.8.tar.gz
All component versions are from recent master branch. Full list of
changes is below.
OpenSM:
===
Alex Netes (1):
opensm: fixed
On 15:59 Sun 13 Feb , Jason Gunthorpe wrote:
On Sun, Feb 13, 2011 at 09:55:50AM +0200, Alex Netes wrote:
We finished the management tree seperation.
From now on, Ira Weiny wei...@llnl.gov takes the responsibility for
maitaining libibmad and
infiniband-diags. His trees
Hi Jason,
On 15:12 Fri 11 Feb , Jason Gunthorpe wrote:
* Properly byteswap block_num before using it
* Properly check bounds of the block number before de-referencing it.
This fixes a remotely triggered null pointer dereference.
---
opensm/include/iba/ib_types.h |2 +-
On 15:32 Mon 14 Feb , Ira Weiny wrote:
From: Ira Weiny wei...@llnl.gov
Date: Mon, 14 Feb 2011 15:25:36 -0800
Subject: [PATCH] Fix compile warning introduced by patch fixed getline
pointer allocation free in osm_console_io
Applied. Thanks.
--
To unsubscribe from this list: send the
On 14:01 Wed 23 Feb , Hal Rosenstock wrote:
On Wed, Feb 23, 2011 at 12:20 PM, Alex Netes ale...@voltaire.com wrote:
Hi Jason,
On 15:12 Fri 11 Feb , Jason Gunthorpe wrote:
* Properly byteswap block_num before using it
* Properly check bounds of the block number before de
On 12:35 Fri 25 Feb , Jason Gunthorpe wrote:
On Fri, Feb 25, 2011 at 01:25:15PM +0200, Alex Netes wrote:
What is the expected behaviour when IB PKey table block is empty?
rec.pkey_tbl might be uninitialized here.
Shouldn't SubnAdmGetResp contain ERR_NO_RECORDS in such case
From e4525b15f7b8c721a2e8bde57db2891c004a18b7 Mon Sep 17 00:00:00 2001
From: Alex Netes ale...@voltaire.com
Date: Mon, 28 Feb 2011 18:27:08 +0200
Subject: [PATCH] opensm: fixed indentation and decreased verbosity of RMPP
length message
Decreased log verbosity of RMPP X length X message to DEBUG
Fix INCS path not to contain double slash, as it cause to failure when
creating RPM package on RedHat 6.
Signed-off-by: Aleksey Senin aleks...@voltaire.com
---
defs.mk |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/defs.mk b/defs.mk
index 578d407..11256c0 100644
---
calculation
Reply-To:
When number of ports participating in a group is less than 2, port_list isn't
freed.
Signed-off-by: Alex Netes ale...@mellanox.com
---
opensm/osm_mcast_mgr.c |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/opensm/osm_mcast_mgr.c b/opensm
There is a scenario that fopen will overwrite output variable.
Signed-off-by: Alex Netes ale...@mellanox.com
---
opensm/osm_console.c |2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/opensm/osm_console.c b/opensm/osm_console.c
index fad80ad..7003508 100644
--- a/opensm
struct torus *nt should be checked for validity before getting assignments.
Signed-off-by: Alex Netes ale...@mellanox.com
---
opensm/osm_torus.c | 10 +++---
1 files changed, 7 insertions(+), 3 deletions(-)
diff --git a/opensm/osm_torus.c b/opensm/osm_torus.c
index add3cf9..7a2c252 100644
Variable temp is unsigned.
Signed-off-by: Alex Netes ale...@mellanox.com
---
opensm/main.c |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/opensm/main.c b/opensm/main.c
index 756fe6f..ef4b805 100644
--- a/opensm/main.c
+++ b/opensm/main.c
@@ -987,7 +987,7 @@ int main
Need to free all the allocated variables in case of error.
Signed-off-by: Alex Netes ale...@mellanox.com
---
opensm/osm_ucast_ftree.c | 26 +-
1 files changed, 21 insertions(+), 5 deletions(-)
diff --git a/opensm/osm_ucast_ftree.c b/opensm/osm_ucast_ftree.c
index
This fix is a cosmetic one, but this is the corect way to define sizeof in
malloc.
Signed-off-by: Alex Netes ale...@mellanox.com
---
opensm/osm_ucast_lash.c |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/opensm/osm_ucast_lash.c b/opensm/osm_ucast_lash.c
index
lru might be uninitialized in case when m-version = ~0.
Signed-off-by: Alex Netes ale...@mellanox.com
---
libvendor/osm_vendor_ibumad.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/libvendor/osm_vendor_ibumad.c b/libvendor/osm_vendor_ibumad.c
index be2dda2..5cd6f46
p_madw might be freed in umad_send(). So printing TID instead of the p_madw
pointer is more accurate.
Signed-off-by: Alex Netes ale...@mellanox.com
---
libvendor/osm_vendor_ibumad.c |8 +---
1 files changed, 5 insertions(+), 3 deletions(-)
diff --git a/libvendor/osm_vendor_ibumad.c b
Hi Jason,
On 17:29 Tue 01 Mar , Jason Gunthorpe wrote:
This value must match the portGUID, so it needs to vary on a per port
basis like portGUID and not simply reflect the value opensm acquired
during the sweep.
Prior to this patch opensm returns the same value for localPortNum for
all
On 10:11 Wed 09 Mar , Jason Gunthorpe wrote:
On Wed, Mar 09, 2011 at 05:47:44PM +0200, Alex Netes wrote:
Hi Jason,
On 17:29 Tue 01 Mar , Jason Gunthorpe wrote:
This value must match the portGUID, so it needs to vary on a per port
basis like portGUID and not simply reflect
On 00:47 Thu 02 Sep , Hal Rosenstock wrote:
Most of what the osmtest SA client is doing is not RMPP.
For example, any SA GetTable request is not an RMPP operation
but is requesting an RMPP response.
SA AttributeOffset only needs to be set in an RMPP DATA packet
so this is not needed
On 20:28 Thu 17 Jun , Hal Rosenstock wrote:
Handle PKey and QPN redirection information
GID redirection handling remains
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
---
Changes since v4:
Fixed some trailing whitespace problems
Changes since v3:
Rebased
Changes since
On 21:16 Thu 17 Jun , Hal Rosenstock wrote:
Follows previous patch that adds better redirection support for PerfMgr
Signed-off-by: Hal Rosenstock hal.rosenst...@gmail.com
---
Changes since v4:
Fixed rejection of Copyright hunk
Changes since v3:
Fixed some formatting problems (spaces
On 17:29 Tue 01 Mar , Jason Gunthorpe wrote:
This value must match the portGUID, so it needs to vary on a per port
basis like portGUID and not simply reflect the value opensm acquired
during the sweep.
Prior to this patch opensm returns the same value for localPortNum for
all ports on a
On 18:39 Thu 06 Jan , Eli Dorfman (Voltaire) wrote:
[PATCH] configure multicast only once during sweep
opensm performed multicast configuration twice during sweep by calling
two different function that are very similar.
remove osm_mcast_mgr_process() which had a bug and didn't clear mlid
Enables to define list of switch ports so the SM will
go over this list when creating a routing.
It helps balancing links load on some communication patterns
where multipile links connect between the switches.
Signed-off-by: David McMillen da...@systemfabricworks.com
Signed-off-by: Alex Netes ale
On 09:16 Sun 20 Mar , Tamir Ronen wrote:
From acc805cdf65e12e2cee9bfbf360a176c55a5949f Mon Sep 17 00:00:00 2001
From: Tamir Ronen tam...@mellanox.com
Date: Thu, 17 Mar 2011 15:21:43 +0200
Subject: [PATCH] Ignore command line option -t 0
If the timeout specified in the command line
Adding Congestion Control definitions as described in IB spec A10
and Link Working Group Errata Q3/2010.
Signed-off-by: Alex Netes ale...@mellanox.com
---
Makefile.am |4 +-
include/infiniband/mad.h | 67 +-
src/ccm.c| 102
On 10:37 Mon 21 Mar , Hefty, Sean wrote:
+#undef DEBUG
+#define DEBUG if (ibdebug)IBWARN
I didn't see where this was used, but in general this sort of definition may
not work. It can incorrectly associate a 'else' class with the if (ibdebug)
statement, rather than the
On 17:25 Mon 07 Mar , Yevgeny Kliteynik wrote:
If there was a syntax error in the QoS policy file,
any subsequent parsing fails, even if the error is fixed.
The reason is that the parser doesn't clean its buffer -
need to clean it explicitly when exiting parsing.
Signed-off-by: Aviad
On 00:04 Sat 26 Mar , Hal Rosenstock wrote:
Rather than setting mfttop to 0xbfff first time, initialize (indicated
by first_time_master_sweep) based on whether or not SA DB is being loaded.
If it's being loaded, get mfttop from there (based on max MLID in the
SA DB). Otherwise, it's set to
On 18:11 Thu 24 Mar , Ira Weiny wrote:
From: Ira Weiny wei...@llnl.gov
Date: Wed, 9 Mar 2011 16:28:23 -0800
Subject: [PATCH] opensm/perfmgr: fix overflow processing
The stored error counters need to be cleared when the clear is issued
because of overflow.
Signed-off-by: Ira
On 10:46 Tue 29 Mar , Eli Dorfman wrote:
Subject: [PATCH] Allow comma in plugin names parsing
Signed-off-by: Eli Dorfman el...@mellanox.com
---
Apllied, thanks.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a message to majord...@vger.kernel.org
Hi Ira,
On 18:20 Thu 24 Mar , Ira Weiny wrote:
I have briefly tested the following 4 patches to implement
PortCountersExtended
collection in the performance manager of opensm. They work well for my
limited
testing.
Can you please provide more info regarding the tests you've run.
Hi Ira,
On 18:20 Thu 24 Mar , Ira Weiny wrote:
I have briefly tested the following 4 patches to implement
PortCountersExtended
collection in the performance manager of opensm. They work well for my
limited
testing.
Can you provide more info regarding the tested you've done?
Hi Hal,
On 14:38 Fri 01 Apr , Hal Rosenstock wrote:
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
Applied, thanks.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a message to majord...@vger.kernel.org
More majordomo info at
On 15:25 Mon 04 Apr , Hal Rosenstock wrote:
rather than error (it's preferred but not required)
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
Applied, thanks.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a message to
Hi Al, Jared,
On 14:31 Wed 23 Mar , Albert Chu wrote:
1) Port Shifting
This is similar to what was done with some of the LMC 0 code.
Congestion would occur due to alignment of routes w/ common traffic
patterns. However, we found that it was also necessary for LMC=0 and
only
Hi Hal,
On 14:05 Tue 05 Apr , Hal Rosenstock wrote:
PKey table capacities are not required to be multiples of the PKey table block
size (32 entries of 16 pkeys).
Current code could enable partition enforcement on the peer switch port
even if the last partition table block were
On 08:45 Tue 05 Apr , Hal Rosenstock wrote:
Changes for not setting attribute offset unless RMPP
Added previous changes not merged:
Added OSMV_QUERY_MULTIPATH_REC support in osmv_query_sa
Changed bind info to indicate both is_responder and is_report_processor
are FALSE rather than TRUE
On 14:04 Tue 05 Apr , Hal Rosenstock wrote:
Closer to where initially used
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
Applied, thanks.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a message to majord...@vger.kernel.org
More majordomo
On 14:05 Tue 05 Apr , Hal Rosenstock wrote:
Invalid status may be returned for the insufficient components tests
Also, set pkey when component mask set
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
Applied, thanks
--
To unsubscribe from this list: send the line unsubscribe
On 14:31 Thu 07 Apr , Hal Rosenstock wrote:
Hi Alex,
On 4/7/2011 12:46 PM, Alex Netes wrote:
Hi Hal,
On 14:05 Tue 05 Apr , Hal Rosenstock wrote:
PKey table capacities are not required to be multiples of the PKey table
block
size (32 entries of 16 pkeys
On 15:02 Thu 07 Apr , Hal Rosenstock wrote:
Checking Create given MGID=0 skip Qkey and Pkey (o15.0.1.4) - Error 02A7
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
Applied, thanks
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a message to
Hi Ken,
On 13:29 Wed 13 Apr , Alex Netes wrote:
-Original Message-
From: Schmidt, Kenneth P [mailto:kenneth.schm...@pnl.gov]
Sent: Monday, March 28, 2011 3:49 AM
To: Alex Netes
Cc: Sasha Khapyorsky; Carr, Jared F
Subject: Re: [PATCH] OpenSM - The DnUp routing algorithm
Hi Ken,
This is the first round of comments.
First of all, I noticed that there is a big amount of code that is almost
identical to UpDn. Have you considered to 'tweak' osm_ucast_updn.c instead of
creating a new file? I'm not sure that it's the best way to go, but it might
reduce the code. The
On 18:23 Tue 22 Mar , Albert Chu wrote:
Hey Jim, Alex,
Just hit a segfault on the main tree. It appears patch
commit 9ddcf3419eade13bdc0a54f93930c49fe67efd63
Author: Jim Schutt jasc...@sandia.gov
Date: Fri Sep 3 10:43:12 2010 -0600
opensm: Avoid havoc in minhop caused by
Hi Ira,
On 15:17 Fri 15 Apr , Ira Weiny wrote:
Subject: [PATCH] opensm: perfmgr, only set orig_lid when we have a valid
port. Otherwise leave it as 0
Signed-off-by: Ira Weiny wei...@llnl.gov
---
opensm/osm_perfmgr.c |4 +++-
1 files changed, 3 insertions(+), 1 deletions(-)
On 15:19 Fri 15 Apr , Ira Weiny wrote:
Subject: [PATCH] opensm: fix strtoull error handling
The man page states that errno _may_ be returned when the string can not be
converted. Check that the entire string was consumed otherwise assume this is
not a guid.
Signed-off-by: Ira Weiny
Hi Hal,
On 11:00 Mon 11 Apr , Hal Rosenstock wrote:
alias GUIDs are those indicated in SM GUIDInfo attribute
(other than index 0) and are created/removed via SA Set/Delete
of GUIDInfoRecord per IBTA MgtWG errata RefIDs 4704-4706.
Up to now, this attribute was unneeded by the SM as
On 08:20 Wed 20 Apr , Hal Rosenstock wrote:
Hi Alex,
On 4/20/2011 8:08 AM, Alex Netes wrote:
Hi Hal,
On 11:00 Mon 11 Apr , Hal Rosenstock wrote:
alias GUIDs are those indicated in SM GUIDInfo attribute
(other than index 0) and are created/removed via SA Set/Delete
Hi Hal,
On 11:00 Mon 11 Apr , Hal Rosenstock wrote:
This change prepares for support of alias GUIDs by adding an alias guid
port table and currently populates it only with the base port GUID
in the response to the SM NodeInfo attribute query.
Ultimately, base as well as alternate
Hi Hal,
On 09:29 Mon 18 Apr , Hal Rosenstock wrote:
For base SP0, PortState in SM PortInfo attribute is not used and base SP0
is always active.
How SM can be attached to base SP0?
Moreover, if during discovery we found that SM is attached to base SP0, don't
we want to act like the port
On 10:43 Wed 20 Apr , Hal Rosenstock wrote:
Hi Alex,
On 4/20/2011 10:17 AM, Alex Netes wrote:
Hi Hal,
On 09:29 Mon 18 Apr , Hal Rosenstock wrote:
For base SP0, PortState in SM PortInfo attribute is not used and base SP0
is always active.
How SM can be attached
On 10:26 Tue 19 Apr , Hal Rosenstock wrote:
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
Applied, thanks.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a message to majord...@vger.kernel.org
More majordomo info at
On 10:27 Tue 19 Apr , Hal Rosenstock wrote:
in sr_rcv_process_delete_method
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
Applied, thanks.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a message to majord...@vger.kernel.org
More majordomo
On 10:27 Tue 19 Apr , Hal Rosenstock wrote:
in sr_rcv_process_delete_method
Also, cosmetic formatting change
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
Applied, thanks.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a message to
Hi Sean,
On 11:21 Thu 21 Apr , Hefty, Sean wrote:
Replace CreateDirectory with SHCreateDirectoryEx, so that
intermediate directories in the database path are created.
This fixes an issue where opensm fails to start unless the
path has been created beforehand.
Signed-off-by: Sean
On 17:45 Tue 26 Apr , Ira Weiny wrote:
From: Ira Weiny wei...@llnl.gov
Date: Tue, 26 Apr 2011 17:35:38 -0700
Subject: [PATCH] opensm: ib_types.h add defines for PM CPI
SamplesOnlySupported and PortCountersXmitWaitSupported
Per IBTA v1.2.1 section 16.1.3.1
Signed-off-by: Ira Weiny
Hi Al,
On 11:02 Wed 20 Apr , Albert Chu wrote:
Do not use the dimn_ports_file if not using dor routing. If the
dimn_ports_file is specified, it currently will be utilized regardless
of the routing algorithm actually being used.
Al
--
I guess that dimn_ports_file option was designed
Hi Hal,
On 15:39 Mon 11 Apr , Hal Rosenstock wrote:
Fixes seg fault when ib_pkey_is_invalid called in last_used_pkey_index
Introduced by commit aebd9f9d9cdd1c60c2b3784c0c03cfe8f4014019
for better last block handling
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
diff --git
On 11:15 Wed 27 Apr , Hal Rosenstock wrote:
Hi Alex,
On 4/27/2011 10:57 AM, Alex Netes wrote:
Hi Hal,
On 15:39 Mon 11 Apr , Hal Rosenstock wrote:
Fixes seg fault when ib_pkey_is_invalid called in last_used_pkey_index
Introduced by commit
Hi Hal,
On 11:00 Mon 11 Apr , Hal Rosenstock wrote:
alias GUIDs are those indicated in SM GUIDInfo attribute
(other than index 0) and are created/removed via SA Set/Delete
of GUIDInfoRecord per IBTA MgtWG errata RefIDs 4704-4706.
Up to now, this attribute was unneeded by the SM as
Hi Hal,
On 11:00 Mon 11 Apr , Hal Rosenstock wrote:
This change prepares for support of alias GUIDs by adding an alias guid
port table and currently populates it only with the base port GUID
in the response to the SM NodeInfo attribute query.
Ultimately, base as well as alternate
Hi Hal,
On 09:29 Mon 18 Apr , Hal Rosenstock wrote:
For base SP0, PortState in SM PortInfo attribute is not used and base SP0
is always active.
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
Applied. Thanks.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma
Hi Hal,
On 09:29 Mon 18 Apr , Hal Rosenstock wrote:
For base SP0, PortState in SM PortInfo attribute is not used and base SP0
is always active.
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
Applied. Thanks.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma
Hi Hal,
On 15:39 Mon 11 Apr , Hal Rosenstock wrote:
In pkey_mgr_update_peer_port, rather than mirror the end port
PKey table, pack the peer switch's port PKey table to
eliminate any holes.
Why don't you want to pack port's pkey table itself and mirror it packed to
the peer's port?
Hi Sean,
I want to make sure I didn't miss any libibumad patchaes. There should be 8
patches:
[PATCH 1/7] libibumad: provide MAD definitions with libibumad
http://www.spinics.net/lists/linux-rdma/msg07997.html
[PATCH 2/7] libibumad: add SA MAD definitions to umad
Enables to define list of switch ports so the SM will
go over this list when creating a routing.
It helps balancing links load on some communication patterns
where multipile links connect between the switches.
Signed-off-by: David McMillen da...@systemfabricworks.com
Signed-off-by: Alex Netes ale
Hi Eli,
On 11:02 Tue 03 May , Eli Dorfman wrote:
clean partition enforcement on inter-switch links
after connectivity change from switch-host to switch-switch the
partition enforcement bits were not cleared on the port.
since the pkey table is not set on inter-switch links,
packets
Hi Hal,
On 09:27 Fri 29 Apr , Hal Rosenstock wrote:
From 28edb24012fe79ed5556dc612ad0d4b4b0d8c571 Mon Sep 17 00:00:00 2001
From: Hal Rosenstock h...@mellanox.com
Date: Thu, 28 Apr 2011 22:48:15 +0300
Subject: [PATCH] opensm: Make SA assigned guids persistent across port
down/up events
MFTSubnSet failed MADs may leave temporary MC loops in the fabric.
In order to eliminate this faulty state as quick as possible it's a good
thing to initiate a heavy sweep immediately and to wait for the next light
sweep.
Signed-off-by: Alex Netes ale...@mellanox.com
---
opensm/osm_state_mgr.c
Hi Hal,
On 09:27 Fri 29 Apr , Hal Rosenstock wrote:
From b587d02d3ab47e3fe47f98b2c8115c7686e6104e Mon Sep 17 00:00:00 2001
From: Hal Rosenstock h...@mellanox.com
Date: Thu, 28 Apr 2011 22:53:34 +0300
Subject: [PATCH] opensm: Dump/load SA GUIDInfoRecords
into port alias guid table
Hi Hal,
Some cosmetic comments.
On 09:28 Fri 29 Apr , Hal Rosenstock wrote:
From 2c0c38ef7294e55833d70c7be721ad51a50fcf06 Mon Sep 17 00:00:00 2001
From: Hal Rosenstock h...@mellanox.com
Date: Fri, 29 Apr 2011 15:02:10 +0300
Subject: [PATCH] opensm: Handle SubnSet GUIDInfo asynchronously
Hi Hal,
On 19:45 Mon 02 May , Hal Rosenstock wrote:
Hi Alex,
On 5/2/2011 8:12 AM, Alex Netes wrote:
Hi Hal,
On 15:39 Mon 11 Apr , Hal Rosenstock wrote:
In pkey_mgr_update_peer_port, rather than mirror the end port
PKey table, pack the peer switch's port PKey table
Hi Hal,
On 10:35 Fri 20 May , Hal Rosenstock wrote:
introduced by commit 2c5148408d562d86e1dd5d90939a58533ddebbe6
to eliminate compiler warning:
osm_pkey_mgr.c: In function 'osm_pkey_mgr_process':
osm_pkey_mgr.c:561: warning: assignment from incompatible pointer type
Signed-off-by:
Hi Hal,
On 14:09 Fri 20 May , Hal Rosenstock wrote:
In pkey_mgr_update_peer_port, rather than mirror the end port
PKey table, pack the peer switch's port PKey table to
eliminate any holes.
This is only done for switch external ports and not end ports
because it's disruptive to change
Hi Hal,
On 15:39 Mon 11 Apr , Hal Rosenstock wrote:
Fixes seg fault when ib_pkey_is_invalid called in last_used_pkey_index
Introduced by commit aebd9f9d9cdd1c60c2b3784c0c03cfe8f4014019
for better last block handling
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
Applied,
Commit 1c2a298b295eba7e24205519abc24e47106d15df broke port order
configuration for torus routing engine. order was incorrectly initiated,
causing setting LFTs to fail.
Signed-off-by: Alex Netes ale...@mellanox.com
---
opensm/osm_torus.c |4 ++--
1 files changed, 2 insertions(+), 2 deletions
...@dev.mellanox.co.il
Signed-off-by: Alex Netes ale...@mellanox.com
---
include/opensm/osm_subnet.h |9 -
opensm/osm_sm_state_mgr.c |1 +
opensm/osm_state_mgr.c | 82 +++
3 files changed, 91 insertions(+), 1 deletions(-)
diff --git a/include/opensm
On 20:40 Thu 02 Jun , Hal Rosenstock wrote:
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
Applied, thanks.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a message to majord...@vger.kernel.org
More majordomo info at
On 20:40 Thu 02 Jun , Hal Rosenstock wrote:
Since port-priv can be NULL, struct osm_remote_guids_count * supplied
can be NULL so handle this.
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
Applied, thanks.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
On 15:03 Thu 02 Jun , Hal Rosenstock wrote:
osm_vendor_mlx.c: In function '__osmv_get_send_txn':
osm_vendor_mlx.c:708: warning: format '%llX' expects type 'long long unsigned
int', but argument 4 has type 'uint64_t'
osm_vendor_mlx.c:723: warning: format '%llX' expects type 'long long
On 16:33 Wed 01 Jun , Jim Schutt wrote:
Since it doesn't cause OpenSM to exit, it isn't an error.
We just want to give the admin information that will allow them
to clean up their configuration.
Signed-off-by: Jim Schutt jasc...@sandia.gov
---
Applied, thanks.
--
To unsubscribe from
On 16:33 Wed 01 Jun , Jim Schutt wrote:
Signed-off-by: Jim Schutt jasc...@sandia.gov
---
Applied, thanks.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a message to majord...@vger.kernel.org
More majordomo info at
Hi Hal,
On 17:21 Fri 22 Apr , Hal Rosenstock wrote:
Default is on; as this is a workaround for non compliance:
this feature is advertised but the SMA rejects sets of SwitchInfo that
actually set MFTTop.
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
Applied, thanks.
--
To
Hi Hal,
On 11:06 Sat 04 Jun , Hal Rosenstock wrote:
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
Applied, thanks.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a message to majord...@vger.kernel.org
More majordomo info at
Hi Al,
On 10:07 Wed 27 Apr , Albert Chu wrote:
Hey Alex,
On Wed, 2011-04-27 at 03:50 -0700, Alex Netes wrote:
Hi Al,
On 11:02 Wed 20 Apr , Albert Chu wrote:
Do not use the dimn_ports_file if not using dor routing. If the
dimn_ports_file is specified, it currently
. It still not optimal solution though.
Maybe more nice approach would be to leave --dimn_ports_file option as is, but
to add more info in the man page and in the usage message.
What do you think?
Al
On Tue, 2011-06-28 at 01:49 -0700, Alex Netes wrote:
Hi Al,
On 10:07 Wed 27 Apr
in v2: leave --dimn_ports_file in opensm.conf
Signed-off-by: Hal Rosenstock h...@mellanox.com
Signed-off-by: Alex Netes ale...@mellanox.com
---
include/opensm/osm_subnet.h |2 +-
include/opensm/osm_switch.h |6 ++--
man/opensm.8.in | 27 ++-
opensm/main.c
Hi Hal,
On 07:33 Tue 05 Jul , Hal Rosenstock wrote:
osm_switch.c: In function 'osm_switch_recommend_path':
osm_switch.c:282: warning: 'found_node_guid' may be used uninitialized in
this function
osm_switch.c:281: warning: 'found_sys_guid' may be used uninitialized in this
function
Hi Hal,
On 07:33 Tue 05 Jul , Hal Rosenstock wrote:
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
Applied, thanks.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a message to majord...@vger.kernel.org
More majordomo info at
Hi Or,
On 12:07 Thu 07 Jul , Or Gerlitz wrote:
Hi Alex,
I just noted that opensm logs can go anywhere in size when one of my
nodes had no space left on device after the opensm log level was raised...
I see that the rpm provided with RHEL6 is setting an
/etc/logrotate.d/opensm
entry
Hi Ira,
On 14:39 Wed 06 Jul , Ira Weiny wrote:
Signed-off-by: Ira Weiny wei...@llnl.gov
---
Applied, thanks.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a message to majord...@vger.kernel.org
More majordomo info at
Hi Ira,
On 15:11 Wed 06 Jul , Ira Weiny wrote:
This should at least be compiled in by default.
Signed-off-by: Ira Weiny wei...@llnl.gov
---
Sounds like a good idea.
If we change the defaults, why should we keep this option for configure
script? Is it really needed?
What about
Hi Hal,
On 13:24 Tue 05 Jul , Hal Rosenstock wrote:
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
Applied, thanks.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a message to majord...@vger.kernel.org
More majordomo info at
Hi Ira,
On 10:13 Fri 08 Jul , Ira Weiny wrote:
On Fri, 8 Jul 2011 09:23:50 -0700
Hal Rosenstock h...@dev.mellanox.co.il wrote:
Ira,
On 7/8/2011 12:06 PM, Ira Weiny wrote:
However, when we first put the PerfMgr in OpenSM there was some concern
for developers who may be using
1 - 100 of 384 matches
Mail list logo