Hi Hal,
-Original Message-
From: Hal Rosenstock [mailto:[EMAIL PROTECTED]
Sent: Wednesday, April 05, 2006 1:59 PM
To: Yael Kalka
Cc: openib-general@openib.org; Eitan Zahavi; Sasha Khapyorsky; Ofer
Gigi
Subject: Re: [PATCH] OpenSM - complib fix for branch
Hi Yael,
On Wed, 2006
Hi Hal,
The patch fixes the problem I saw.
Please apply it.
Thanks,
Yael
-Original Message-
From: Hal Rosenstock [mailto:[EMAIL PROTECTED]
Sent: Wednesday, April 05, 2006 3:55 PM
To: Yael Kalka
Cc: openib-general@openib.org
Subject: [PATCHv2] OpenSM: Fix osm_vendor_send for GSI
Hi Hal,
I saw that the complib patch (removal of constructor and destructor
attribute), wasn't fully added to the branch.
Attached is a patch for the branch.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: complib/cl_complib.c
the rmpp header.
Please apply the patch on both trunk and branch 1.0.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: libvendor/osm_vendor_ibumad.c
===
--- libvendor/osm_vendor_ibumad.c (revision 6192
PM
To: Yael Kalka
Cc: openib-general@openib.org; Eitan Zahavi; Sasha Khapyorsky; Ofer
Gigi
Subject: Re: [PATCH] OpenSM - fix osm_vendor_send on vendor mads
Hi Yael,
On Wed, 2006-04-05 at 04:54, Yael Kalka wrote:
Hi Hal,
We saw the following problem in the osm_vendor_send mad
patch changes this to rate 120GB/sec.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: osmtest/osmt_multicast.c
===
--- osmtest/osmt_multicast.c(revision 5972)
+++ osmtest/osmt_multicast.c(working copy
Hi Hal,
Attached is the patch for adding guid to opensm opts for the 1.0 branch.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: include/opensm/osm_subnet.h
===
--- include/opensm/osm_subnet.h (revision 5955
Hi Hal,
Attached is the patch for adding honor_guid2lid flag for the 1.0 branch.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: include/opensm/osm_subnet.h
===
--- include/opensm/osm_subnet.h (revision 5955
-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_sa_class_port_info.c
===
--- opensm/osm_sa_class_port_info.c (revision 5919)
+++ opensm/osm_sa_class_port_info.c (working copy)
@@ -69,7 +69,7 @@
#define MAX_MSECS_TO_RTV 24
Hi Hal,
Attached is another fix for windows - again some casting issue.
And also a comment naming fix in ib_types.h.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: include/iba/ib_types.h
===
--- include/iba
) - then it shouldn't issue a
duplicated guids error (and exit), but just ignore and continue.
This means that only if the SM recognizes such a duplication in a
stable subnet it'll issue the error and exit.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_node_info_rcv.c
), instead of checking the sm_key in the
request mad.
The following patch fixes this issue, and also removes a comment out
of place.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_sa_mcmember_record.c
.
This patch enables to run opensm with --honor_guid2lid flag, and that
forces the opensm to re-read the guid2lid file when it comes out of
Standby, and if this file is legal - honor it.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: include/opensm/osm_subnet.h
.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_lid_mgr.c
===
--- opensm/osm_lid_mgr.c(revision 5753)
+++ opensm/osm_lid_mgr.c(working copy)
@@ -900,10 +900,11
Hi Hal,
From your mail, I understand that you haven't received this patch I
sent you.
Here it is again.
Thanks,
Yael
Yael Kalka wrote:
Attached is a patch removing complib_fini from cl_complib.c, which is
duplicated in complib_exit.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED
Hi Hal,
Attached is a patch removing complib_fini from cl_complib.c, which is
duplicated in complib_exit.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: complib/cl_complib.c
===
--- complib/cl_complib.c
Hi Hal,
Patch sent separately.
Thanks,
Yael
-Original Message-
From: Hal Rosenstock [mailto:[EMAIL PROTECTED]
Sent: Thursday, March 09, 2006 2:25 PM
To: Yael Kalka
Cc: openib-general@openib.org; Eitan Zahavi; Sasha Khapyorsky; Ofer
Gigi
Subject: Re: [PATCH] OpenSM - complib
Hi Hal,
The following patch adds the full path name of the guid2lid file in
case of an error in parsing the file (which causes opensm to exit).
Also - there is a check that the guid given is in legal size.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_db_files.c
in opensm/main.c, as
main.c is different between linux and windows stacks.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: complib/cl_complib.c
===
--- complib/cl_complib.c(revision 5671)
+++ complib/cl_complib.c
,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: include/opensm/osm_sa_mcmember_record.h
===
--- include/opensm/osm_sa_mcmember_record.h (revision 5608)
+++ include/opensm/osm_sa_mcmember_record.h (working copy
, and fail, since the port capabilities
are not realizable.
The following patch adds creation of the MC group with MTU(rate) that
also takes into account the MTU(rate) of the port requesting the creation.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: include/opensm
Hi Hal,
-Original Message-
From: [EMAIL PROTECTED] [mailto:openib-general-
[EMAIL PROTECTED] On Behalf Of Hal Rosenstock
Sent: Tuesday, February 28, 2006 3:39 PM
To: Yael Kalka
Cc: openib-general@openib.org
Subject: [openib-general] Re: [PATCH] OpenSM
Hi Sasha,
The patch compiles fine under windows.
Thanks,
Yael
-Original Message-
From: Sasha Khapyorsky [mailto:[EMAIL PROTECTED]
Sent: Wednesday, February 22, 2006 1:33 AM
To: [EMAIL PROTECTED]; Yael Kalka
Cc: openib-general@openib.org
Subject: [PATCH] opensm: fixes in signal
Hi Hal,
Currently osm_vendor_get_all_port_attr doesn't update the port number
information. The following patch adds this information.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: libvendor/osm_vendor_ibumad.c
- checks
of creation with illegal rate and mtu.
Also in the patch is a change to osm_helper.c, osm_dump_mc_record -
dump all the records of mc (previously not all were printed).
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: osmtest/osmt_multicast.c
Hello Sasha,
There are problems with the patch you sent - doesn't work when trying to
apply it.
One problem, for example - is that file opensm/osm_partition.h is given
with diffs, when actually it is a new file.
Yael
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]
To: Yael Kalka
Cc: openib-general@openib.org; Eitan Zahavi
Subject: Re: Re:[PATCH] OpenSM/st.c: Fix some size_t issues related
tomemoryallocation in st.c
Hi Yael,
On Sun, 2006-02-19 at 07:24, Yael Kalka wrote:
Hi Hal,
The patch in general is fine. I've added one change to the original
patch
Hi Hal,
Here is the patch with the fix for the type of i.
patch - to avoid casting issues underwindows
Below is the full patch.
Yael
Signed-off-by: Hal Rosenstock [EMAIL PROTECTED]
Index: include/opensm/st.h
===
---
Hi Hal,
The patch in general is fine. I've added one change to the original
patch - to avoid casting issues underwindows
Below is the full patch.
Yael
Signed-off-by: Hal Rosenstock [EMAIL PROTECTED]
Index: include/opensm/st.h
Hi Hal,
The following patch fixes some problems in osmt_service.c flow,
along with cosmetic cleanups of the code.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: osmtest/osmt_service.c
===
--- osmtest
Hi Hal,
I am re-sending the patch with another fix. Please ignore this patch.
Thanks,
Yael
-Original Message-
From: Yael Kalka
Sent: Wednesday, February 15, 2006 12:07 PM
To: [EMAIL PROTECTED]
Cc: openib-general@openib.org; Eitan Zahavi; Yael Kalka
Subject: [PATCH] Opensm
Hi Hal,
The following patch add the following to osmt_service.c:
1. Currently, the flow sometimes exits with TEST PATH although the test
actually fails.
2. Added cleanup of all the services created at the end of the test.
3. Cosmetic cleanups of the code.
Thanks,
Yael
Signed-off-by: Yael
Hi Hal,
The following patch adds includes changes to support windows
compilation. The changes are of files to include, and a casting fix.
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: osmtest/osmtest.c
===
--- osmtest
Hi Hal,
The following patch removes an extra log message from osmtest.c
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: osmtest/osmtest.c
===
--- osmtest/osmtest.c (revision 5380)
+++ osmtest/osmtest.c
Hi Hal,
The following patch changes some of the includes for the windows compilation.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: include/opensm/osm_log.h
===
--- include/opensm/osm_log.h(revision 5380
Hi Hal,
The following patch changes some of the includes for the windows
compilation.
Also there are fixes of type casting.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: osmtest/osmt_service.c
===
--- osmtest
Hi Hal,
Yet another patch with changes of the some includes for the windows
compilation, and type casting fixes.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: osmtest/osmt_slvl_vl_arb.c
===
--- osmtest
Hi Hal,
The following patch cleans up the osm_ucast_updn.c construct function
that you pointed out, and also adds a check if the construct succeeded
during updn_init.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_ucast_updn.c
09, 2006 9:09 PM
To: Yael Kalka
Cc: openib-general@openib.org; Eitan Zahavi
Subject: Re: [PATCH] Opensm - clean osm_vendor_mlx_sa.c code
Hi Yael,
On Mon, 2006-02-06 at 07:39, Yael Kalka wrote:
Hi Hal,
Currently in osm_vendor_mlx_sa.c the sent context is saved arbitrarily
as nodeInfo_context
To: Yael Kalka
Cc: openib-general@openib.org; Eitan Zahavi
Subject: Re: [PATCH] Opensm - cl_event_wheel casting
Hi Yael,
On Mon, 2006-02-06 at 03:53, Yael Kalka wrote:
Hi Hal,
The following patch adds the casting done in a clearer way - to avoid
compilation errors in windows. Also - added
merging, for example).
The following patch adds support for turning on ClientReRegistration
bit on newly discovered ports.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_lid_mgr.c
===
--- opensm
Hi Hal,
The following patch adds a missing type casting in the return value of
the function osm_mcast_mgr_compute_max_hops.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: osm_mcast_mgr.c
===
--- osm_mcast_mgr.c
Hi Hal,
There was a problem with some of the types defined when compiling on
64bit windows machines. The following patch adds support for these as
well.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: include/opensm/st.h
Hi Hal,
The following patch includes some fixes for the windows stack:
1. Add needed __cdecl.
2. Change the default directories/files names.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: include/opensm/osm_base.h
Hi Hal,
The Patch Michael Tsirkin suggested for fixing the OSM_LOG_ENTER
problem works fine both for windows and for linux.
Here is the patch for this, instead of the previous one I sent.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: include/opensm/osm_log.h
Hi Hal,
The following patch adds the casting done in a clearer way - to avoid
compilation errors in windows. Also - added a clear message if the
timeout was trimmed (due to the casting).
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: complib/cl_event_wheel.c
Hi Hal,
The following patch adds some changes in osm_db_file.c to match the
windows stack.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_db_files.c
===
--- opensm/osm_db_files.c (revision 5307
Hi Hal,
The following patch adds some missing casts and fixes object types to
fix compilation errors in the windows stack, aadds some changes in
osm_db_file.c to match the windows stack.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_db_pack.c
is to add parenthesis on the code handling the
multicast path records. This patch adds these parenthesis.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_sa_path_record.c
===
--- opensm/osm_sa_path_record.c
Hi Hal,
The original static allocation doesn't compile in Windows.
The attached patch replaces it with dynamic allocation.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_ucast_mgr.c
===
--- opensm
Hi Hal,
The osm_reg_sig_handler function is not supported in Windows.
The following patch adds the function only if non-Windows stack.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: include/opensm/osm_opensm.h
: Monday, February 06, 2006 2:25 PM
To: Yael Kalka
Cc: [EMAIL PROTECTED]; openib-general@openib.org
Subject: Re: [PATCH] Opensm - osm_reg_sig_handler in Windows
Quoting r. Yael Kalka [EMAIL PROTECTED]:
Subject: [PATCH] Opensm - osm_reg_sig_handler in Windows
Hi Hal,
The osm_reg_sig_handler
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: libvendor/osm_vendor_mlx_sa.c
===
--- libvendor/osm_vendor_mlx_sa.c (revision 5307)
+++ libvendor/osm_vendor_mlx_sa.c (working copy)
@@ -96,9 +96,9
Hi Hal,
Currently SYSLOG prints are not executed under Windows.
The following patch adds these printings to the Windows stack as well.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_log.c
Hi Hal,
Currently the build of opensm fails, since osm_svn_revision.h is
missing (and created only later).
The following patch issued by Michael Tsirkin fixes it.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/Makefile.am
The Following patch removes these asserts.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_pkey_rcv.c
===
--- opensm/osm_pkey_rcv.c (revision 5246
Hi Hal,
I saw that there is a problem with the printing of the error message
in __osm_pi_rcv_process_switch_port - the link state printed is the
state saved on the port, and not the state checked in the portInfo
received.
The attached patch fixes this.
Thanks,
Yael
Signed-off-by: Yael Kalka
Hi Hal,
The following patch adds the use of the OSM_DEFAULT_CACHE_DIR instead
of full name usage. Also - add the / at the end of the
OSM_DEFAULT_CACHE_DIR definition, and refrain from adding it in the
code, to avoid problems in Windows.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED
Hi Hal,
Looks good. Go ahead and add it.
Thanks,
Yael
-Original Message-
From: Hal Rosenstock [mailto:[EMAIL PROTECTED]
Sent: Saturday, January 28, 2006 7:23 PM
To: Yael Kalka; Eitan Zahavi
Cc: Troy Benjegerdes; Troy Benjegerdes; openib-general@openib.org
Subject: [PATCHv2] OpenSM
.
This was the problem left.
Using O_NONBLOCK works fine for me. I will send a patch seperately with this
fix instead of the original one.
Thanks,
Yael
-Original Message-
From: Hal Rosenstock [mailto:[EMAIL PROTECTED]
Sent: Thursday, January 19, 2006 4:51 PM
To: Yael Kalka
Cc: openib-general
Hi Hal,
Here is a patch using O_NONBLOCK instead of the static variable.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: libvendor/osm_vendor_ibumad.c
===
--- libvendor/osm_vendor_ibumad.c (revision 4951
Hi Hal,
I've noticed that when running opensm with --console, it exits with a
message that option `-console' requires an argument. But I saw in the
main.c that such argument isn't used.
The following patch removes this dependency.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED
to
open /dev/inifiniband/issm%d file twice during the run of opensm.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: libvendor/osm_vendor_ibumad.c
===
--- libvendor/osm_vendor_ibumad.c (revision 4951)
+++ libvendor
Hi Hal,
You are correct.
Thanks,
Yael
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] Behalf Of Hal Rosenstock
Sent: Monday, January 09, 2006 5:39 PM
To: Yael Kalka
Cc: openib-general@openib.org
Subject: [openib-general] Re: Re[PATCH] Opensm - fix
: Monday, January 09, 2006 4:56 PM
To: Yael Kalka
Cc: openib-general@openib.org; Eitan Zahavi
Subject: Re: Re[PATCH] Opensm - clean exit on ^C
Hi Yael,
On Thu, 2006-01-05 at 07:16, Yael Kalka wrote:
Hi Hal,
I've noticed that sometimes when killing OpenSM using ^C not all
threads are killed
Hi Hal,
During some tests we've notices that not all compmask fields are
properly checked and compared in the portInfo record query.
Attached is a patch with the missing checks, and addition of some
set/get relevant functions added to the ib_types.h as well.
Thanks,
Yael
Signed-off-by: Yael
0, we get 5 ports instead of 4,
and the third port (which was the data copied as the default port for
the second hca) is not valid.
The following patch fixes this.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: libvendor/osm_vendor_ibumad.c
patch fixes this.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: include/complib/cl_signal_osd.h
===
--- include/complib/cl_signal_osd.h (revision 4760)
+++ include/complib/cl_signal_osd.h (working copy
seen errors due to this is because the main and
usual flow is fine.
I will issue a patch for fixing this soon.
Thanks,
Yael
-Original Message-
From: Eitan Zahavi
Sent: Saturday, December 31, 2005 11:43 AM
To: 'Hal Rosenstock'; Yael Kalka
Cc: openib-general@openib.org
Subject: RE: Some
Hi Hal,
Attached is a fix for the qp0_mads_outstanding handling, according to
what I've described in the previous mail.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_vl15intf.c
===
--- opensm
-general@openib.org
Subject: RE: [openib-general] RE: [PATCH] Opensm - fix segfault on exit
- cont.
Hi Greg,
On Tue, Dec 20, 2005 at 09:11:12AM +0200, Yael Kalka wrote:
- if (p_ur-signal)
+ if (p_ur-signal != NULL)
Aren't these 2 statements required to execute the same
this.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: libvendor/osm_vendor_ibumad.c
===
--- libvendor/osm_vendor_ibumad.c (revision 4542)
+++ libvendor/osm_vendor_ibumad.c (working copy)
@@ -552,7 +552,7
Hi Hal,
I've noticed that under certain operating systems, when driver isn't
loaded, the SM still exits with segfault.
The following patch fixes this.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: libvendor/osm_vendor_ibumad.c
Hi Hal,
Currently, if the user level driver installation is not in /usr/local/
dir - configure of opensm failes.
The following patch enables support for arbitrary paths of user level
driver installation. Path can be given using --with-uldrv flag.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL
Hi Hal,
Hi Yael,
On Thu, 2005-12-08 at 05:39, Yael Kalka wrote:
Hi Hal,
If osm_vendor_get_all_port_attr is called before the osm_vendor_bind,
What exercises the vendor calls in this manner ?
[YK] - We saw it in some test that uses the vendor lib. Later on the
bind was called
Hi Hal,
During some tests here, we noticed that if the SA is queried with
IB_PIR_COMPMASK_BASELID, and base_lid = zero - the SA will return in
result all the ports.
The following patch fixes this.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_sa_portinfo_record.c
Hi Hal,
If osm_vendor_get_all_port_attr is called before the osm_vendor_bind,
then the sm_lid of the default port isn't updated correctly.
This patch fixes it.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: libvendor/osm_vendor_ibumad.c
This looks good.
Please apply it.
Thanks,
Yael
-Original Message-
From: Hal Rosenstock [mailto:[EMAIL PROTECTED]
Sent: Wednesday, December 07, 2005 2:06 PM
To: Yael Kalka; Eitan Zahavi
Cc: openib-general@openib.org
Subject: [PATCH] OpenSM: SubnAdmGet PathRecord should assume NumbPath of
1
the patch...
This issue should be added to our to-do list.
Yael
-Original Message-
From: Hal Rosenstock [mailto:[EMAIL PROTECTED]
Sent: Tuesday, December 06, 2005 4:06 PM
To: Yael Kalka
Cc: openib-general@openib.org
Subject: [PATCH] OpenSM: SA SMInfoRecord should support GetTable as well
[mailto:[EMAIL PROTECTED]
Sent: Wednesday, December 07, 2005 2:15 PM
To: Yael Kalka
Cc: openib-general@openib.org
Subject: RE: [PATCH] OpenSM: SA SMInfoRecord should support GetTable as
well as Get method
Hi Yael,
On Wed, 2005-12-07 at 06:59, Yael Kalka wrote:
Hi Hal,
If you look at the code
Hi Hal,
The following code exists at least in several of the osm_sa_*_record.c
files, but is missing in the osm_sa_node_record.c.
When running with debug level - add a dump of the node record sent
in the request.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm
Didn't look at all of them, but I know at least some of them include the
record dumping.
-Original Message-
From: Hal Rosenstock [mailto:[EMAIL PROTECTED]
Sent: Tuesday, December 06, 2005 2:15 PM
To: Yael Kalka
Cc: openib-general@openib.org; Eitan Zahavi
Subject: Re: [PATCH] Opensm
Hi Hal,
If the driver isn't loaded, opensm exits with segfault. This is since
it tries destroying the signal event in the osm_vendor, but due to the
failure - this event wasn't created.
The following patch fixes this.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: libvendor
- there is no need to update anything in the SMInfo SubnGet
request.
Any objections?
Yael
-Original Message-
From: Hal Rosenstock [mailto:[EMAIL PROTECTED]
Sent: Thursday, December 01, 2005 7:02 PM
To: Eitan Zahavi
Cc: Yael Kalka; openib-general@openib.org
Subject: RE: [Fwd: [openib-general] OpenSM
handling of the issue - first,
issue an error message to the /var/log/messeges file as well.
Second - add an option flag to the SM that will define wether or not
to exit on such case.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: include/opensm/osm_subnet.h
: Wednesday, November 30, 2005 11:57 PM
To: Yael Kalka; Eitan Zahavi
Cc: openib-general@openib.org
Subject: [Fwd: [openib-general] OpenSM and Wrong SM_Key]
Hi Yael Eitan,
Based on the recent MgtWG discussions, are you still holding your
position in terms of exiting OpenSM when a non matching SM Key
-Original Message-
From: Hal Rosenstock [mailto:[EMAIL PROTECTED]
Sent: Wednesday, November 30, 2005 4:20 PM
To: Yael Kalka
Cc: openib-general@openib.org; Eitan Zahavi
Subject: Re: [PATCH] Opensm - fix LinkRecord get
On Wed, 2005-11-30 at 07:35, Yael Kalka wrote:
Hi Hal,
During some tests I've
Good catch...
-Original Message-
From: Hal Rosenstock [mailto:[EMAIL PROTECTED]
Sent: Tuesday, November 29, 2005 5:12 PM
To: Yael Kalka
Cc: openib-general@openib.org; Eitan Zahavi
Subject: [PATCH] [TRIVIAL] osm_node_info_rcv.c:osm_ni_rcv_init Move
assert before variable used
During some tests I've noticed that in PathRecord queries the SA
doesn't compare the packetLifeTime component, if relevant comp_mask is
turned on. This patch fixes this.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_sa_path_record.c
is actually different
than null. As a result we can get several links with the same value.
This patch fixes both issues.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_sa_link_record.c
===
--- opensm
Hi Hal,
This fix isn't correct, since you are asserting on a variable not yet
initialized.
Yael
-Original Message-
From: Hal Rosenstock [mailto:[EMAIL PROTECTED]
Sent: Wednesday, November 30, 2005 11:04 PM
To: Yael Kalka
Cc: openib-general@openib.org
Subject: [PATCH] [TRIVIAL] OpenSM
Currently the error message generated if there are duplicated guids on
the subnet is not clear, and is devided to several error messages.
The following patch fixes the error message to be clearer.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_node_info_rcv.c
Hi
Hal,
LMC
0 has been tested, though not heavily, on theOpenIB
stack.
Yael
-Original Message-From: Hal Rosenstock
[mailto:[EMAIL PROTECTED]Sent: Monday, November 28, 2005 3:58
AMTo: Robert WalshCc:
openib-general@openib.orgSubject: Re: [openib-general]
on the client_reregsitration issue next
week, and fix this issue as well.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_lid_mgr.c
===
--- opensm/osm_lid_mgr.c(revision 4119)
+++ opensm/osm_lid_mgr.c
Hal Rosenstock wrote:
On Wed, 2005-11-23 at 04:37, Yael Kalka wrote:
Hi Hal,
Currently when sending mads with PortInfo Set we set client_rereg to
be zero
I think you mean one here.
You are right.
if we are in first_time_master_sweep and relevant capability
mask is on). On other
(and not TRUE), as these are all responses.
In osm_sm_mad_ctrl - we can send both requests and responses (for
SMInfo, for example), so no use in checking the resp_expected flag.
This patch fixes these issues.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: opensm/osm_sa_mad_ctrl.c
Hi Hal,
You are correct. The Notices are requests. Then the assertion on the
resp_expected
should be removed from the osm_sa_mad_ctrl.c.
Thanks,
Yael
-Original Message-
From: Hal Rosenstock [mailto:[EMAIL PROTECTED]
Sent: Tuesday, November 22, 2005 3:14 PM
To: Yael Kalka
Cc: openib
Hi Hal,
While reviewing the umad_receiver function in osm_vendor_ibumad.c we've
noticed
that when umad_alloc() calls fail, the receiver breaks.
What happens then is that SM continues to live, though the umad_receiver
thread
doesn't exist anymore.
I think that there is no use in keeping the SM
Hi Hal,
We are encountering problems in umad_send with large sized mads. I
will send a different mail regarding this issue to the group.
This patch adds the mad size to the error message when umad_send
failed.
Thanks,
Yael
Signed-off-by: Yael Kalka [EMAIL PROTECTED]
Index: libvendor
Hello all,
During some opensm tests we've encountered a situation where a call
to umad_send() fails on very large mads (on mads of size ~90,000).
Is there some limitation on the size of the mads that can be sent?
Reproduction of the issue is simple:
Run opensm
Then run osmtest -f m -M 3
What the
1 - 100 of 153 matches
Mail list logo