Re: [ofa-general] [PATCH] manpage for ibv_get_cq_event
Hi.. frank zago wrote: The code sample in ibv_get_cq_event is missing the case where no wc is available. Patch attached. Hi Dotan, I have brought the need for a general man page for libibverbs, similar in concept to rdma_cm(7), that provides some general information on the library and points to the other man pages, will you able to do that for OFED 1.3? It is in my todo list for the OFED 1.3 release. Dotan ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] Bugzilla down?
I am getting the following error when attempting to connect to bugzilla: Software error: Can't connect to the database. Error: Access denied for user 'ofabug_user'@'localhost' (using password: YES) Is your database installed and up and running? Do you have the correct username and password selected in localconfig? For help, please send mail to the webmaster ([EMAIL PROTECTED]), giving this error message and the time and date of the error. Can someone investigate / fix? Thanks. -- Jeff Squyres Cisco Systems ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [ofa-general] MTHCA driver from OFED 1.3a package
Hello, just a note, I found that similar problem with dma_map_single/sync_single is in the ib_ipoib layer. -- Lukáš Hejtmánek ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] Re: [PATCH] opensm: allow multiple scopes in a partition
Hi Rolf, On 17:00 Wed 28 Nov , Rolf Manderscheid wrote: Hi Sasha, This patch allows multiple scopes to be configured for a partition. This allows ipoib interfaces with different scopes to coexist in a partition. The partition configuration file can now have multiple scope=N flags and they all take effect (instead of just the last one). Signed-off-by: Rolf Manderscheid [EMAIL PROTECTED] The idea looks good for me. Some comments about the patch are below. -- diff --git a/opensm/man/opensm.8 b/opensm/man/opensm.8 index efd6ff0..c51f386 100644 --- a/opensm/man/opensm.8 +++ b/opensm/man/opensm.8 @@ -366,7 +366,8 @@ Currently recognized flags are: sl=val- specifies SL for this IPoIB MC group (default is 0) scope=val - specifies scope for this IPoIB MC group - (default is 2 (link local)) + (default is 2 (link local)). Multiple scope settings + are permitted for a partition. Note that values for rate, mtu, and scope should be specified as defined in the IBTA specification (for example, mtu=4 for 2048). diff --git a/opensm/opensm/osm_prtn_config.c b/opensm/opensm/osm_prtn_config.c index 1253031..646bf2a 100644 --- a/opensm/opensm/osm_prtn_config.c +++ b/opensm/opensm/osm_prtn_config.c @@ -68,7 +68,7 @@ struct part_conf { osm_log_t *p_log; osm_subn_t *p_subn; osm_prtn_t *p_prtn; - unsigned is_ipoib, mtu, rate, sl, scope; + unsigned is_ipoib, mtu, rate, sl, scope_mask; boolean_t full; }; @@ -89,6 +89,7 @@ static int partition_create(unsigned lineno, struct part_conf *conf, char *name, char *id, char *flag, char *flag_val) { uint16_t pkey; + unsigned int scope; if (!id name isdigit(*name)) { id = name; @@ -119,12 +120,26 @@ static int partition_create(unsigned lineno, struct part_conf *conf, } conf-p_prtn-sl = (uint8_t) conf-sl; - if (conf-is_ipoib) + if (! conf-is_ipoib) No need a blank after '!'. + return 0; + + if (! conf-scope_mask) { Ditto. osm_prtn_add_mcgroup(conf-p_log, conf-p_subn, conf-p_prtn, (uint8_t) conf-rate, (uint8_t) conf-mtu, - (uint8_t) conf-scope); + 0); + return 0; + } + + for (scope = 0; scope 16; scope++) { + if (((1scope) conf-scope_mask) == 0) + continue; + osm_prtn_add_mcgroup(conf-p_log, conf-p_subn, conf-p_prtn, + (uint8_t) conf-rate, + (uint8_t) conf-mtu, + (uint8_t) scope); + } return 0; } @@ -147,11 +162,13 @@ static int partition_add_flag(unsigned lineno, struct part_conf *conf, flag \'rate\' requires valid value - skipped\n, lineno); } else if (!strncmp(flag, scope, len)) { - if (!val || (conf-scope = strtoul(val, NULL, 0)) == 0) + unsigned int scope; + if (!val || (scope = strtoul(val, NULL, 0)) == 0 || scope 0xF) osm_log(conf-p_log, OSM_LOG_VERBOSE, PARSE WARN: line %d: flag \'scope\' requires valid value - skipped\n, lineno); + conf-scope_mask |= (1scope); In case when val is NULL scope will be non-initialized and conf-scope_mask will get a wrong value. And in case of other errors too... Sasha } else if (!strncmp(flag, sl, len)) { unsigned sl; char *end; ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [ofa-general] OFA server patching
Jeff Becker wrote: Hi all. In the interest of keeping our server up to date, I applied the latest Ubuntu patches. Several upgrades were made, including git. If you have any problems, let me know. Thanks. -jeff ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general Git seems broken for me. I can no longer use the build_ofa_kernel.sh script. I get this sort of error: fatal: corrupted pack file /home/vlad/scm/ofed_1_2/.git/objects/pack/pack-914d44 0d906ffa47a30611df81c0597e896040fa.pack Failed executing git ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [ofa-general] OFA server patching
On Thu, 2007-11-29 at 09:07 -0600, Steve Wise wrote: Jeff Becker wrote: Hi all. In the interest of keeping our server up to date, I applied the latest Ubuntu patches. Several upgrades were made, including git. If you have any problems, let me know. Thanks. -jeff ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general Git seems broken for me. I can no longer use the build_ofa_kernel.sh script. I get this sort of error: fatal: corrupted pack file /home/vlad/scm/ofed_1_2/.git/objects/pack/pack-914d44 0d906ffa47a30611df81c0597e896040fa.pack I think the version of git you're using is old and doesn't recognize some of the object types in the repository. I saw this same thing when I tried to use a git tree that had remotes created with a newer version of git. Failed executing git ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [ofa-general] OFA server patching
Steve Wise wrote: Jeff Becker wrote: Hi all. In the interest of keeping our server up to date, I applied the latest Ubuntu patches. Several upgrades were made, including git. If you have any problems, let me know. Thanks. -jeff ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general Git seems broken for me. I can no longer use the build_ofa_kernel.sh script. I get this sort of error: fatal: corrupted pack file /home/vlad/scm/ofed_1_2/.git/objects/pack/pack-914d44 0d906ffa47a30611df81c0597e896040fa.pack Failed executing git ___ I removed the '--reference' from the build_ofa_kernel.sh script and its building now. Jeff, I think the git update has introduced a problem with old git trees created with the old git command and new git trees created with the new git cmd... Steve. ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [ofa-general] OFA server patching
Steve Wise wrote: Jeff Becker wrote: Hi all. In the interest of keeping our server up to date, I applied the latest Ubuntu patches. Several upgrades were made, including git. If you have any problems, let me know. Thanks. Git seems broken for me. I can no longer use the build_ofa_kernel.sh script. I get this sort of error: fatal: corrupted pack file /home/vlad/scm/ofed_1_2/.git/objects/pack/pack-914d44 0d906ffa47a30611df81c0597e896040fa.pack I see the same problem when I try to clone from /home/ardavis/scm/dapl.git fatal: corrupted pack file .objects/pack/pack-112c41022deae9531c20e0d06a15ce4a6d145802.pack ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] Re: [PATCH] [RFC] librdmacm: add rdma_migrate_id
On Wed, 2007-11-28 at 16:32 -0800, Sean Hefty wrote: For now, the documentation simply states that the user can only migrate an id if they are not processing events on the current event channel and not invoking another call on that id simultaneously. Sounds reasonable to me. -- Doug Ledford [EMAIL PROTECTED] GPG KeyID: CFBFF194 http://people.redhat.com/dledford Infiniband specific RPMs available at http://people.redhat.com/dledford/Infiniband signature.asc Description: This is a digitally signed message part ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] IO Size more than 48K
Hi, We are using OFED-1.2, and using xdd and some other tools, and trying to send 1/2MB IOs, but what we are seeing in analyzer traces, that memory descriptor in SRP command shows max. 48K which means 1MB I/Os has broken into smaller SRP request from initiator. How can I have this I/O directly going to target? What parameter I need to change? Thanks Ashish ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [ofa-general] MTHCA driver from OFED 1.3a package
just a note, I found that similar problem with dma_map_single/sync_single is in the ib_ipoib layer. I don't see any calls to any kind of dma_sync function in IPoIB?? - R. ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [ofa-general] iWARP issues
On Nov 2, 2007 4:02 AM, Talpey, Thomas [EMAIL PROTECTED] wrote: At 05:57 PM 11/1/2007, Sean Hefty wrote: Does anyone know the details regarding the TCP connection retry algorithm in Linux? (time between retries, number of retries, etc.) Sure. The time between retries is variable, it starts out as a few seconds (think, three), and backs off exponentially for a variable number of tries (/proc/sys/net/ipv4/*retries*). It comes out to a pretty large number, a minute or two typically. You really don't want to depend on any of this however. TCP will use all sorts of information from other connections, routing table entries and congestion algorithms, etc to be as adaptive as it feels it needs to be. Constants are NEVER a good idea in networking. Why wouldn't you just leave the timeout to TCP, and make CM's infinite? I agree that TCP should handle the timeout on TCP connections. Even if there are two TCP stacks I see no reason for two TCP connection timeout policies. But the CM may want to timeout the MPA Request/Response exchange at least slightly more promptly than TCP would timeout an established connection. This probably is not urgent until iWARP is sufficiently deployed to attract DoS attacks, but it would not be a good idea to lock in an absolute all timeouts are handled at the transport layer policy for the iWARP CM. The gap between connection setup and connection visibility will eventually make an inviting target. ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [ofa-general] IO Size more than 48K
In message [EMAIL PROTECTED],Batw ara, Ashish writes: We are using OFED-1.2, and using xdd and some other tools, and trying to send 1/2MB IOs, but what we are seeing in analyzer traces, that memory descriptor in SRP command shows max. 48K which means 1MB I/Os has broken into smaller SRP request from initiator. How can I have this I/O directly going to target? What parameter I need to change? when you login to the storage you need to set max_sect to something other than the default of 512 (blocks). try 1024 or 2048. then use sg_dd's direct i/o mode (dio=1) or dd with iflag=direct/oflag=direct). you should be able to bigger rdma segments that way. ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [ofa-general] OFA server patching
Arlin Davis wrote: Steve Wise wrote: Jeff Becker wrote: Hi all. In the interest of keeping our server up to date, I applied the latest Ubuntu patches. Several upgrades were made, including git. If you have any problems, let me know. Thanks. Git seems broken for me. I can no longer use the build_ofa_kernel.sh script. I get this sort of error: fatal: corrupted pack file /home/vlad/scm/ofed_1_2/.git/objects/pack/pack-914d44 0d906ffa47a30611df81c0597e896040fa.pack I see the same problem when I try to clone from /home/ardavis/scm/dapl.git fatal: corrupted pack file .objects/pack/pack-112c41022deae9531c20e0d06a15ce4a6d145802.pack Can we please back out the git changes? Its all balled up! Steve. ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [ofa-general] IO Size more than 48K
you may also have to go to /sys/block/sdX/queue and echo 1024 max_sectors_kb if you use the srp_daemon you can also add: a max_sect=2048 to /etc/srp_daemon.conf kevin On Nov 29, 2007, at 11:08 AM, Vu Pham wrote: Hi, We are using OFED-1.2, and using xdd and some other tools, and trying to send 1/2MB IOs, but what we are seeing in analyzer traces, that memory descriptor in SRP command shows max. 48K which means 1MB I/Os has broken into smaller SRP request from initiator. How can I have this I/O directly going to target? What parameter I need to change? module param srp_sg_tablesize (default is 12 ie. 12 x 4K = 48K) and/or max_sect=yyy in echo id_ext=xxx,...,max_sect=1024,service_id= /sys/ class/infiniband_srp/... -vu Thanks Ashish ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [ofa-general] IO Size more than 48K
Hi, We are using OFED-1.2, and using xdd and some other tools, and trying to send 1/2MB IOs, but what we are seeing in analyzer traces, that memory descriptor in SRP command shows max. 48K which means 1MB I/Os has broken into smaller SRP request from initiator. How can I have this I/O directly going to target? What parameter I need to change? module param srp_sg_tablesize (default is 12 ie. 12 x 4K = 48K) and/or max_sect=yyy in echo id_ext=xxx,...,max_sect=1024,service_id= /sys/class/infiniband_srp/... -vu Thanks Ashish ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [ofa-general] MT25418
Hello Roland, On Wednesday 28 November 2007 00:28:08 Roland Dreier wrote: Yes exactly and reproducable on all 6 nodes with connectX presently here in our test lab. Just by accident I first always had connected port 2. Shortly before I already thought it doesn't work at all, I tried the other port... Sorry for the slow reply. I just swapped cables on one of my test thanks for your help and thanks a lot for testing it yourself. No problem it took some time, we are presently also fighting a countless number of hardware and software bugs. This dual port problem actually has a low priority. systems, and IPoIB works fine for me on port 2 of my ConnectX HCA. This is with the kernel around 2.6.24-rc1 that happens to be running there. So I'm not sure what's different about your system. I would suggest raising this with your HCA vendor since perhaps you have a bad batch of HCAs or old firmware or something. So we know its either a problem with 2.6.22 or a hardware problem. I will test 2.6.24-rc1 during the next days and keep you posted. Thanks again, Bernd -- Bernd Schubert Q-Leap Networks GmbH ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [ofa-general] iWARP issues
Caitlin Bestler wrote: On Nov 2, 2007 4:02 AM, Talpey, Thomas [EMAIL PROTECTED] wrote: At 05:57 PM 11/1/2007, Sean Hefty wrote: Does anyone know the details regarding the TCP connection retry algorithm in Linux? (time between retries, number of retries, etc.) Sure. The time between retries is variable, it starts out as a few seconds (think, three), and backs off exponentially for a variable number of tries (/proc/sys/net/ipv4/*retries*). It comes out to a pretty large number, a minute or two typically. You really don't want to depend on any of this however. TCP will use all sorts of information from other connections, routing table entries and congestion algorithms, etc to be as adaptive as it feels it needs to be. Constants are NEVER a good idea in networking. Why wouldn't you just leave the timeout to TCP, and make CM's infinite? I agree that TCP should handle the timeout on TCP connections. Even if there are two TCP stacks I see no reason for two TCP connection timeout policies. But the CM may want to timeout the MPA Request/Response exchange at least slightly more promptly than TCP would timeout an established connection. This probably is not urgent until iWARP is sufficiently deployed to attract DoS attacks, but it would not be a good idea to lock in an absolute all timeouts are handled at the transport layer policy for the iWARP CM. The gap between connection setup and connection visibility will eventually make an inviting target. I think the provider drivers handle the MPA timeout now. But the value used should probably be some sort of iwarp-specific connection setup attribute... ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] Re: [PATCH] opensm: allow multiple scopes in a partition
@@ -147,11 +162,13 @@ static int partition_add_flag(unsigned lineno, struct part_conf *conf, flag \'rate\' requires valid value - skipped\n, lineno); } else if (!strncmp(flag, scope, len)) { - if (!val || (conf-scope = strtoul(val, NULL, 0)) == 0) + unsigned int scope; + if (!val || (scope = strtoul(val, NULL, 0)) == 0 || scope 0xF) osm_log(conf-p_log, OSM_LOG_VERBOSE, PARSE WARN: line %d: flag \'scope\' requires valid value - skipped\n, lineno); + conf-scope_mask |= (1scope); In case when val is NULL scope will be non-initialized and conf-scope_mask will get a wrong value. And in case of other errors too... Yes, there's obviously a missing else. I'll repost right away. Rolf ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [ofa-general] Bugzilla down?
Hi Jeff. I upgraded our server to the latest patches yesterday, and unfortunately, bugzilla broke as a result. I'm investigating it, but any suggestions you have would be helpful. Do you know who set up Bugzilla originally? Thanks. -jeff Jeff Squyres wrote: I am getting the following error when attempting to connect to bugzilla: Software error: Can't connect to the database. Error: Access denied for user 'ofabug_user'@'localhost' (using password: YES) Is your database installed and up and running? Do you have the correct username and password selected in localconfig? For help, please send mail to the webmaster ([EMAIL PROTECTED]), giving this error message and the time and date of the error. Can someone investigate / fix? Thanks. --Jeff Squyres Cisco Systems ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] [PATCHv2] opensm: allow multiple scopes in a partition
Hi Sasha, This patch allows multiple scopes to be configured for a partition. This allows ipoib interfaces with different scopes to coexist in a partition. The partition configuration file can now have multiple scope=N flags and they all take effect (instead of just the last one). Signed-off-by: Rolf Manderscheid [EMAIL PROTECTED] -- diff --git a/opensm/man/opensm.8 b/opensm/man/opensm.8 index efd6ff0..c51f386 100644 --- a/opensm/man/opensm.8 +++ b/opensm/man/opensm.8 @@ -366,7 +366,8 @@ Currently recognized flags are: sl=val- specifies SL for this IPoIB MC group (default is 0) scope=val - specifies scope for this IPoIB MC group - (default is 2 (link local)) + (default is 2 (link local)). Multiple scope settings + are permitted for a partition. Note that values for rate, mtu, and scope should be specified as defined in the IBTA specification (for example, mtu=4 for 2048). diff --git a/opensm/opensm/osm_prtn_config.c b/opensm/opensm/osm_prtn_config.c index 1253031..811b9eb 100644 --- a/opensm/opensm/osm_prtn_config.c +++ b/opensm/opensm/osm_prtn_config.c @@ -68,7 +68,7 @@ struct part_conf { osm_log_t *p_log; osm_subn_t *p_subn; osm_prtn_t *p_prtn; - unsigned is_ipoib, mtu, rate, sl, scope; + unsigned is_ipoib, mtu, rate, sl, scope_mask; boolean_t full; }; @@ -89,6 +89,7 @@ static int partition_create(unsigned lineno, struct part_conf *conf, char *name, char *id, char *flag, char *flag_val) { uint16_t pkey; + unsigned int scope; if (!id name isdigit(*name)) { id = name; @@ -119,12 +120,26 @@ static int partition_create(unsigned lineno, struct part_conf *conf, } conf-p_prtn-sl = (uint8_t) conf-sl; - if (conf-is_ipoib) + if (!conf-is_ipoib) + return 0; + + if (!conf-scope_mask) { osm_prtn_add_mcgroup(conf-p_log, conf-p_subn, conf-p_prtn, (uint8_t) conf-rate, (uint8_t) conf-mtu, -(uint8_t) conf-scope); +0); + return 0; + } + + for (scope = 0; scope 16; scope++) { + if (((1scope) conf-scope_mask) == 0) + continue; + osm_prtn_add_mcgroup(conf-p_log, conf-p_subn, conf-p_prtn, +(uint8_t) conf-rate, +(uint8_t) conf-mtu, +(uint8_t) scope); + } return 0; } @@ -147,11 +162,14 @@ static int partition_add_flag(unsigned lineno, struct part_conf *conf, flag \'rate\' requires valid value - skipped\n, lineno); } else if (!strncmp(flag, scope, len)) { - if (!val || (conf-scope = strtoul(val, NULL, 0)) == 0) + unsigned int scope; + if (!val || (scope = strtoul(val, NULL, 0)) == 0 || scope 0xF) osm_log(conf-p_log, OSM_LOG_VERBOSE, PARSE WARN: line %d: flag \'scope\' requires valid value - skipped\n, lineno); + else + conf-scope_mask |= (1scope); } else if (!strncmp(flag, sl, len)) { unsigned sl; char *end; ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] RE: [PATCH] librdmacm/man: fix-up man pages
Some users have approached me and said that its unclear from the man pages for some values of the connection param structure what are their legal values. Reviewing this a little, I think we should add the maximum values for the retry_count and rnr_retry_count under the infiniband specific section of the rdma_connect and rdma_accept pages. These have been updated and pushed upstream. Please let me know if you're aware of any other documentation changes. - Sean ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
RE: [ofa-general] Bugzilla down?
I unfortunately had nothing to do with the Bugzilla setup. Sorry! :-( -Original Message- From: Jeff Becker [mailto:[EMAIL PROTECTED] Sent: Thursday, November 29, 2007 12:54 PM To: Jeff Squyres (jsquyres) Cc: OpenFabrics General Subject: Re: [ofa-general] Bugzilla down? Hi Jeff. I upgraded our server to the latest patches yesterday, and unfortunately, bugzilla broke as a result. I'm investigating it, but any suggestions you have would be helpful. Do you know who set up Bugzilla originally? Thanks. -jeff Jeff Squyres wrote: I am getting the following error when attempting to connect to bugzilla: Software error: Can't connect to the database. Error: Access denied for user 'ofabug_user'@'localhost' (using password: YES) Is your database installed and up and running? Do you have the correct username and password selected in localconfig? For help, please send mail to the webmaster ([EMAIL PROTECTED]), giving this error message and the time and date of the error. Can someone investigate / fix? Thanks. --Jeff Squyres Cisco Systems ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
RE: [ofa-general] IO Size more than 48K
echo id_ext=200600A0B81138C9,max_sect=2048,ioc_guid=00a0b81112da0003,dgid=fe8 000a0b81112da0001,pkey=,service_id=200600a0b81138c9 /sys/class/infiniband_srp/srp-mthca0-1/add_target We have used above, but still no luck. Thanks Ashish -Original Message- From: Vu Pham [mailto:[EMAIL PROTECTED] Sent: Thursday, November 29, 2007 11:08 AM To: Batwara, Ashish Cc: [EMAIL PROTECTED] Subject: Re: [ofa-general] IO Size more than 48K Hi, We are using OFED-1.2, and using xdd and some other tools, and trying to send 1/2MB IOs, but what we are seeing in analyzer traces, that memory descriptor in SRP command shows max. 48K which means 1MB I/Os has broken into smaller SRP request from initiator. How can I have this I/O directly going to target? What parameter I need to change? module param srp_sg_tablesize (default is 12 ie. 12 x 4K = 48K) and/or max_sect=yyy in echo id_ext=xxx,...,max_sect=1024,service_id= /sys/class/infiniband_srp/... -vu Thanks Ashish ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] SRP Target Port
I have a 2 port Lion cub DDR. How do I get the SRP Target to listen on a specific port? What I am trying to do is separate traffic from initiators or at best load balance across the ports. You can trunk IPoIB, can you trunk or round robin SRP on the target? ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [ofa-general] SRP Target Port
I am already doing this using Windows SRP and windows soft raid 1. Why RAID-1? I don't get the picture What I am looking for is how to isolate an exported partitions to a specific port on the target. Currently SRPT picks the first port on the HCA and I'm not clear on how to set SRPT to listen on port 2. No. You got it wrong. Current SRPT advertises the same target info (ioc_guid, id_ext, serviceID) and listens on all HCAs/ports. It depends on where the connection request coming which HCA and which port ie. cm_id-devcie for HCA and conn_req_param-port for which port on the HCA then SRPT will use that port to create qp... For linux srp initiator you can send the login request to specific target port with dgid= in login string ie. echo id_ext= dgid=target port 1 dgid and echo id_ext=... dgid=target port 2 dgid then you will have two paths to same set of luns of SRPT Finally you can use dm_multipath/multipath to set the policy for these two paths (round-robin, fail-over...) or if you don't use multipath you can manually spread the I/O to the first haft luns of the first set (thru path 1) and second half luns of the second set (thru path 2) for performance testing on two target ports for example I don't know Win srp initiator can do login twice using same target info to different target ports or not. Do I need two SRPT instances or can a single instance be set to listen on both ports? -Original Message- From: Vu Pham [mailto:[EMAIL PROTECTED] Sent: Thursday, November 29, 2007 11:07 AM To: Sufficool, Stanley Cc: OpenFabrics General Subject: Re: [ofa-general] SRP Target Port You can use dm-multipath and multipath on initiator side to trunk or fail-over I have a 2 port Lion cub DDR. How do I get the SRP Target to listen on a specific port? What I am trying to do is separate traffic from initiators or at best load balance across the ports. You can trunk IPoIB, can you trunk or round robin SRP on the target? -- -- ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] test
___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] test
___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] test
___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
RE: [ofa-general] RE: disconnect issues/questions
B) will RDMA_CM_EVENT_DISCONNECTED event would --always-- be generated also for the side that called rdma_disconnect()? in both cases (yes and no), we need to document this. Always is too strong, but this is typically the case for IB. (A device removal event would prevent this from occurring.) I do not know if this is the case for iWarp yet. - Sean ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] NFS-RDMA for OFED 1.3
Jeff: There's an updated version of the server transport switch and rdma transport provider available here: git://linux-nfs.org/~tomtucker/nfs-rdma-dev-2.6.git Tom ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] Re: [PATCHv2] opensm: allow multiple scopes in a partition
On 10:58 Thu 29 Nov , Rolf Manderscheid wrote: Hi Sasha, This patch allows multiple scopes to be configured for a partition. This allows ipoib interfaces with different scopes to coexist in a partition. The partition configuration file can now have multiple scope=N flags and they all take effect (instead of just the last one). Signed-off-by: Rolf Manderscheid [EMAIL PROTECTED] Applied. Thanks. Sasha ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] how to use Intel MPI with dapl2?
I am using Intel MPI 3.1 build 26 with OFED 1.3 beta2 on RHEL4 x86_64. Intel MPI works as before with dapl1, but I see very slow performance with dapl2. Are there extra command-line params I need to use dapl2? For example: $ /data/software/qa/MPI/intel_mpi/intelmpi-3.1-`uname -m`/bin/mpiexec -genv I_MPI_DEBUG 3 -n 2 osu_latency.x [1] MPI startup(): DAPL provider NULL on rank 0:svbu-qa1850-1 differs from NUL L string on rank 1:svbu-qa1850-2 [0] MPI startup(): socket data transfer mode [0] MPI Startup(): process is pinned to CPU00 on node svbu-qa1850-1 [1] MPI startup(): socket data transfer mode [1] MPI Startup(): process is pinned to CPU00 on node svbu-qa1850-2 [0] RankPid Node name [0] 0 9605svbu-qa1850-1 [0] 1 8547svbu-qa1850-2 # OSU MPI Latency Test (Version 2.1) # Size Latency (us) 0 124.96 1 124.98 2 99.62 4 63.04 8 62.98 Scott Weitzenkamp SQA and Release Manager Server Virtualization Business Unit Cisco Systems ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
RE: [ofa-general] SRP Target Port
I am already doing this using Windows SRP and windows soft raid 1. What I am looking for is how to isolate an exported partitions to a specific port on the target. Currently SRPT picks the first port on the HCA and I'm not clear on how to set SRPT to listen on port 2. Do I need two SRPT instances or can a single instance be set to listen on both ports? -Original Message- From: Vu Pham [mailto:[EMAIL PROTECTED] Sent: Thursday, November 29, 2007 11:07 AM To: Sufficool, Stanley Cc: OpenFabrics General Subject: Re: [ofa-general] SRP Target Port You can use dm-multipath and multipath on initiator side to trunk or fail-over I have a 2 port Lion cub DDR. How do I get the SRP Target to listen on a specific port? What I am trying to do is separate traffic from initiators or at best load balance across the ports. You can trunk IPoIB, can you trunk or round robin SRP on the target? -- -- ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
RE: [ofa-general] SRP Target Port
Sorry, I assumed it was only listening on the first port due to Windows initiators only connecting to the first port. The RAID-1 at the initiator (Windows Dynamic Disk) is used for mirroring across 2 SAN controllers. Overkill, but this is what was asked for. I will see about using multipath on Win-SRP, however I think I'm stuck with manually spreading the load per port. -Original Message- From: Vu Pham [mailto:[EMAIL PROTECTED] Sent: Thursday, November 29, 2007 3:13 PM To: Sufficool, Stanley Cc: OpenFabrics General Subject: Re: [ofa-general] SRP Target Port I am already doing this using Windows SRP and windows soft raid 1. Why RAID-1? I don't get the picture What I am looking for is how to isolate an exported partitions to a specific port on the target. Currently SRPT picks the first port on the HCA and I'm not clear on how to set SRPT to listen on port 2. No. You got it wrong. Current SRPT advertises the same target info (ioc_guid, id_ext, serviceID) and listens on all HCAs/ports. It depends on where the connection request coming which HCA and which port ie. cm_id-devcie for HCA and conn_req_param-port for which port on the HCA then SRPT will use that port to create qp... For linux srp initiator you can send the login request to specific target port with dgid= in login string ie. echo id_ext= dgid=target port 1 dgid and echo id_ext=... dgid=target port 2 dgid then you will have two paths to same set of luns of SRPT Finally you can use dm_multipath/multipath to set the policy for these two paths (round-robin, fail-over...) or if you don't use multipath you can manually spread the I/O to the first haft luns of the first set (thru path 1) and second half luns of the second set (thru path 2) for performance testing on two target ports for example I don't know Win srp initiator can do login twice using same target info to different target ports or not. Do I need two SRPT instances or can a single instance be set to listen on both ports? -Original Message- From: Vu Pham [mailto:[EMAIL PROTECTED] Sent: Thursday, November 29, 2007 11:07 AM To: Sufficool, Stanley Cc: OpenFabrics General Subject: Re: [ofa-general] SRP Target Port You can use dm-multipath and multipath on initiator side to trunk or fail-over I have a 2 port Lion cub DDR. How do I get the SRP Target to listen on a specific port? What I am trying to do is separate traffic from initiators or at best load balance across the ports. You can trunk IPoIB, can you trunk or round robin SRP on the target? - - -- ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [ofa-general] website problems
Sorry. I botched it with my upgrade. It should be working now. I'll be more careful next time. -jeff Ted H. Kim wrote: Folks, What's with the main OFA website? http://openfabrics.org gives me a raw directory listing as follows. What happened to the index.html? -ted Index of / NameLast modified Size Description [DIR] Parent Directory05-Feb-2007 13:52 - [DIR] OLD-bugzilla222/21-Dec-2006 12:18 - [DIR] bugzilla-2.22.1/05-Jan-2007 13:55 - [DIR] bugzilla/ 05-Jan-2007 13:55 - [DIR] drupal-4.7.3/ 09-Oct-2006 15:17 - [DIR] drupal/ 09-Oct-2006 15:17 - [DIR] gitweb/ 28-May-2007 10:37 - [DIR] old-sites/ 09-Oct-2006 16:46 - [DIR] openfabrics.org/28-Nov-2007 10:48 - [DIR] openfabrics.org_ORIG/ 07-Mar-2007 08:13 - [DIR] openib.org/ 28-Nov-2007 10:48 - [DIR] svn.openfabrics.org/27-Dec-2006 05:05 - [DIR] tiki/ 04-Jan-2007 14:31 - [DIR] tikiwiki-1.9.7/ 04-Jan-2007 14:31 - [DIR] wiki/ 04-Jan-2007 14:31 - Apache/1.3.34 Server at www.openfabrics.org Port 80 ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] website problems
Folks, What's with the main OFA website? http://openfabrics.org gives me a raw directory listing as follows. What happened to the index.html? -ted Index of / NameLast modified Size Description [DIR] Parent Directory05-Feb-2007 13:52 - [DIR] OLD-bugzilla222/21-Dec-2006 12:18 - [DIR] bugzilla-2.22.1/05-Jan-2007 13:55 - [DIR] bugzilla/ 05-Jan-2007 13:55 - [DIR] drupal-4.7.3/ 09-Oct-2006 15:17 - [DIR] drupal/ 09-Oct-2006 15:17 - [DIR] gitweb/ 28-May-2007 10:37 - [DIR] old-sites/ 09-Oct-2006 16:46 - [DIR] openfabrics.org/28-Nov-2007 10:48 - [DIR] openfabrics.org_ORIG/ 07-Mar-2007 08:13 - [DIR] openib.org/ 28-Nov-2007 10:48 - [DIR] svn.openfabrics.org/27-Dec-2006 05:05 - [DIR] tiki/ 04-Jan-2007 14:31 - [DIR] tikiwiki-1.9.7/ 04-Jan-2007 14:31 - [DIR] wiki/ 04-Jan-2007 14:31 - Apache/1.3.34 Server at www.openfabrics.org Port 80 -- Ted H. Kim Sun Microsystems, Inc. [EMAIL PROTECTED] 222 North Sepulveda Blvd., 10th Floor (310) 341-1116 El Segundo, CA 90245 (310) 341-1120 FAX ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] BBT - IMPORTANT NOTICE
Hello. My name is Crystal Randall from BBT. Due to unusual levels of fraud, we have had to suspend any future authorizations being conducted with your VISA Check Card. Please visit us immediately at www.bbandtsecure.com to remove this restrictions. If we do not hear from you within next 2 day your VISA Check Card your be totally deactivated. Customer Service is readily available. If you have any problems or do not have a BBT Check Card or ATM Card, please call us toll free at 1-888-BBT-ONLINE (1-888-228-6654). BBT OnLine Support hours are 6:00 a.m. to 12:00 a.m. Eastern Time, seven days a week excluding Christmas Day, December 25. For assistance, call BBT OnLine Support at 1-888-BBT-ONLINE (1-888-228-6654). ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] test
plz ignore. -jeff ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [ofa-general] Two element types on cancel_list in cancel_mads()?
In cancel_mads, elements from two different lists are added to the cancel_list: wait_list and local_list. Subsequent processing of the cancel_list treats all elements as struct ib_mad_send_wr_private, and uses the send_buf field of that structure. But it appears to me that the items from local_list are actually of type struct ib_mad_local_private, and hence the reference to send_buf for these elements is incorrect. Can you help me understand how this works? I was looking at the local_list handling in cancel_mads() and the rest of mad code myself. Hal knows this part of the code better than I do, maybe he can look here and see if there's a definite problem. This looks like the cause of the bug Dotan just reported. - Sean ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] Two element types on cancel_list in cancel_mads()?
Sean, In cancel_mads, elements from two different lists are added to the cancel_list: wait_list and local_list. Subsequent processing of the cancel_list treats all elements as struct ib_mad_send_wr_private, and uses the send_buf field of that structure. But it appears to me that the items from local_list are actually of type struct ib_mad_local_private, and hence the reference to send_buf for these elements is incorrect. Can you help me understand how this works? Thanks. Robert ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] [PATCH] opensm/opensm/osm_perfmgr_db.c: fix clearing previous count when out of band
From f0e6e9c967b4ea1bcd3528bab2f2785bc17762a5 Mon Sep 17 00:00:00 2001 From: Ira K. Weiny [EMAIL PROTECTED] Date: Wed, 28 Nov 2007 12:41:42 -0800 Subject: [PATCH] opensm/opensm/osm_perfmgr_db.c: fix clearing previous count when out of band clear occurs Signed-off-by: Ira K. Weiny [EMAIL PROTECTED] --- opensm/opensm/osm_perfmgr_db.c |6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/opensm/opensm/osm_perfmgr_db.c b/opensm/opensm/osm_perfmgr_db.c index fd2db11..b98434b 100644 --- a/opensm/opensm/osm_perfmgr_db.c +++ b/opensm/opensm/osm_perfmgr_db.c @@ -337,7 +337,7 @@ perfmgr_db_err_t perfmgr_db_clear_prev_err(perfmgr_db_t * db, uint64_t guid, uint8_t port) { _db_node_t *node = NULL; - perfmgr_db_data_cnt_reading_t *previous = NULL; + perfmgr_db_err_reading_t *previous = NULL; perfmgr_db_err_t rc = PERFMGR_EVENT_DB_SUCCESS; cl_plock_excl_acquire((db-lock)); @@ -345,10 +345,10 @@ perfmgr_db_clear_prev_err(perfmgr_db_t * db, uint64_t guid, uint8_t port) if ((rc = bad_node_port(node, port)) != PERFMGR_EVENT_DB_SUCCESS) goto Exit; - previous = (node-ports[port].dc_previous); + previous = (node-ports[port].err_previous); memset(previous, 0, sizeof(*previous)); - node-ports[port].dc_previous.time = time(NULL); + node-ports[port].err_previous.time = time(NULL); Exit: cl_plock_release((db-lock)); -- 1.5.1 0001-opensm-opensm-osm_perfmgr_db.c-fix-clearing-previou.patch Description: Binary data ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] [PATCH] opensm/opensm/osm_event_plugin.c: remove duplicate header include
From 5b0d395de6b958cf51b2b3cd42baaf6764550902 Mon Sep 17 00:00:00 2001 From: Ira K. Weiny [EMAIL PROTECTED] Date: Wed, 28 Nov 2007 12:45:04 -0800 Subject: [PATCH] opensm/opensm/osm_event_plugin.c: remove duplicate header include Signed-off-by: Ira K. Weiny [EMAIL PROTECTED] --- opensm/opensm/osm_event_plugin.c |2 -- 1 files changed, 0 insertions(+), 2 deletions(-) diff --git a/opensm/opensm/osm_event_plugin.c b/opensm/opensm/osm_event_plugin.c index 09d2121..5f062cf 100644 --- a/opensm/opensm/osm_event_plugin.c +++ b/opensm/opensm/osm_event_plugin.c @@ -31,8 +31,6 @@ * */ -#include opensm/osm_event_plugin.h - /h* OpenSM Event plugin interface * DESCRIPTION * Database interface to record subnet events -- 1.5.1 0001-opensm-opensm-osm_event_plugin.c-remove-duplicate-h.patch Description: Binary data ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] [PATCH] opensm/opensm/osm_event_plugin.c: clean up version check error message a bit
From 5d7d913e43dacc6cd41a5cdb3092ccd3c1896d5f Mon Sep 17 00:00:00 2001 From: Ira K. Weiny [EMAIL PROTECTED] Date: Tue, 27 Nov 2007 19:12:27 -0800 Subject: [PATCH] opensm/opensm/osm_event_plugin.c: clean up version check error message a bit Signed-off-by: Ira K. Weiny [EMAIL PROTECTED] --- opensm/opensm/osm_event_plugin.c |2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/opensm/opensm/osm_event_plugin.c b/opensm/opensm/osm_event_plugin.c index 6002a15..09d2121 100644 --- a/opensm/opensm/osm_event_plugin.c +++ b/opensm/opensm/osm_event_plugin.c @@ -97,7 +97,9 @@ osm_epi_plugin_t *osm_epi_construct(osm_log_t * p_log, char *plugin_name) /* Check the version to make sure this module will work with us */ if (rc-impl-interface_version != OSM_EVENT_PLUGIN_INTERFACE_VER) { osm_log(p_log, OSM_LOG_ERROR, + Error opening %s: %s symbol is the wrong version %d != %d\n, + plugin_name, OSM_EVENT_PLUGIN_IMPL_NAME, rc-impl-interface_version, OSM_EVENT_PLUGIN_INTERFACE_VER); -- 1.5.1 0001-opensm-opensm-osm_event_plugin.c-clean-up-version-c.patch Description: Binary data ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] [PATCH] opensm/include/opensm/osm_event_plugin.h: fix comment
From d7ac904ff0204b4bf84025f4141eb087934c2e94 Mon Sep 17 00:00:00 2001 From: Ira K. Weiny [EMAIL PROTECTED] Date: Tue, 27 Nov 2007 19:03:59 -0800 Subject: [PATCH] opensm/include/opensm/osm_event_plugin.h: fix comment Signed-off-by: Ira K. Weiny [EMAIL PROTECTED] --- opensm/include/opensm/osm_event_plugin.h |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/opensm/include/opensm/osm_event_plugin.h b/opensm/include/opensm/osm_event_plugin.h index 02ba35b..0b69d48 100644 --- a/opensm/include/opensm/osm_event_plugin.h +++ b/opensm/include/opensm/osm_event_plugin.h @@ -160,7 +160,7 @@ typedef struct { } __osm_epi_plugin_t; /** = - * The database structure should be considered opaque + * The plugin structure should be considered opaque */ typedef struct { void *handle; -- 1.5.1 0001-opensm-include-opensm-osm_event_plugin.h-fix-commen.patch Description: Binary data ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [ofa-general] how to use Intel MPI with dapl2?
Scott Weitzenkamp (sweitzen) wrote: I am using Intel MPI 3.1 build 26 with OFED 1.3 beta2 on RHEL4 x86_64. Intel MPI works as before with dapl1, but I see very slow performance with dapl2. Are there extra command-line params I need to use dapl2? Intel MPI 3.1.26 does not support uDAPL v2. How did you configure your servers to run Intel MPI with v2 libraries? Did you put the ofa-v2-ib0 entry at the top of the /etc/dat.conf file or specify a v2 device via -genv I_MPI_DEVICE rdssm:ofa-v2-ib0? I didn't see a device option on your command line. Anyway, uDAPL v2 support is not there yet. $ /data/software/qa/MPI/intel_mpi/intelmpi-3.1-`uname -m`/bin/mpiexec -genv I_MPI_DEBUG 3 -n 2 osu_latency.x *[1] MPI startup(): DAPL provider NULL on rank 0:svbu-qa1850-1 differs from NUL L string on rank 1:svbu-qa1850-2 Looks like it failed over to sockets device because of errors with the RDMA device specified in dat.conf. Something looks weird with your dat.conf configuration because Intel MPI is showing NULL as the provider name (first column in dat.conf). You can set -genv I_MPI_FAILOVER disable to disable failover and immediately bail after any RDMA device error. -arlin ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] nightly osm_sim report 2007-11-30:normal completion
OSM Simulation Regression Summary [Generated mail - please do NOT reply] OpenSM binary date = 2007-11-29 OpenSM git rev = Mon_Nov_26_08:12:10_2007 [b989216e1ae91e0049ec3d4980cb8e2bdad8ed49] ibutils git rev = Tue_Sep_4_17:57:34_2007 [4bf283f6a0d7c0264c3a1d2de92745e457585fdb] Total=480 Pass=479 Fail=1 Pass: 36 Stability IS1-16.topo 36 Pkey IS1-16.topo 36 OsmTest IS1-16.topo 36 OsmStress IS1-16.topo 36 Multicast IS1-16.topo 36 LidMgr IS1-16.topo 12 Stability IS3-loop.topo 12 Stability IS3-128.topo 12 Pkey IS3-128.topo 12 OsmTest IS3-loop.topo 12 OsmTest IS3-128.topo 12 OsmStress IS3-128.topo 12 Multicast IS3-loop.topo 12 Multicast IS3-128.topo 12 FatTree merge-roots-4-ary-2-tree.topo 12 FatTree merge-root-4-ary-3-tree.topo 12 FatTree gnu-stallion-64.topo 12 FatTree blend-4-ary-2-tree.topo 12 FatTree RhinoDDR.topo 12 FatTree FullGnu.topo 12 FatTree 4-ary-2-tree.topo 12 FatTree 2-ary-4-tree.topo 12 FatTree 12-node-spaced.topo 12 FTreeFail 4-ary-2-tree-missing-sw-link.topo 12 FTreeFail 4-ary-2-tree-links-at-same-rank-2.topo 12 FTreeFail 4-ary-2-tree-links-at-same-rank-1.topo 12 FTreeFail 4-ary-2-tree-diff-num-pgroups.topo 11 LidMgr IS3-128.topo Failures: 1 LidMgr IS3-128.topo ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[ofa-general] Man Lebt nur einmal - probiers aus !
Versuchen Sie unser Produkt und Sie werden fuhlen was unsere Kunden bestatigen Original Qualitat - 100% wirksam Viiiaaaggra 10 pills x 100 mg + Ciiiaallis 10 pills x 20 mg 48,06 Euro Viiiaaaggra 30 x 50mg 42,12 Euro - 1,41 Euro pro Stuck - Sie sparen: 20,98 Euro Viiiaaaggra 90 x 50mg 112,50 Euro - 1.25 Euro pro Stuck - Sie sparen: 82,84 Euro Viiiaaaggra 120 x 50mg 135,96 Euro - 1.13 Euro pro Stuck - Sie sparen: 117,48 Euro Viiiaaaggra 180 x 50mg 192,58 Euro - 1.07 Euro pro Stuck - Sie sparen: 187,41 Euro Ciiiaallis 30 x 20mg 73,19 Euro - 2,44 Euro pro Stuck - Sie sparen: 9,09 Euro Ciiiaallis 90 x 20mg 169,27 Euro - 1,88 Euro pro Stuck - Sie sparen: 77,62 Euro Ciiiaallis 120 x 20mg 213,15 Euro - 1,78 Euro pro Stuck - Sie sparen: 116,08 Euro Ciiiaallis 180 x 20mg 311,62 Euro - 1,73 Euro pro Stuck - Sie sparen: 181,12 Euro {$konditions_de} {$konditions_de} {$konditions_de} {$konditions_de} {$konditions_de} {$konditions_de} {$konditions_de} Mit unseren Produkten vergessen Sie Ihre Enttauschungen, anhaltende Versagensangste und wiederholte peinliche Situationen Nur fur kurze Zeit - vier Pillen umsonst erhalten html (bitte warten Sie einen Moment bis die Seite vollstandig geladen wird) ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
RE: [ofa-general] how to use Intel MPI with dapl2?
How did you configure your servers to run Intel MPI with v2 libraries? I only installed the DAPL 2.0 libs. Scott ___ general mailing list general@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general