[ewg] [ANNOUNCE] management tarballs release
Hi, There is a new release of the management (OpenSM and infiniband diagnostics) tarballs available in: http://www.openfabrics.org/downloads/management/ md5sum: 5a77490e7f604043f5f902e4e4f5d056 libibumad-1.3.3.tar.gz 5f2615fc152cc6dacdca694c16594351 libibmad-1.3.3.tar.gz 0a50e041932b4abf240c8d4a519e59a4 opensm-3.3.3.tar.gz 2454d7f1103bfaa8692c759332639c3c infiniband-diags-1.5.3.tar.gz All component versions are from recent master branch. Full change log is below. Sasha Al Chu (9): libibnetdisc cleanup patches fix libibnetdisc API consistency and bugs fix libibnetdisc API consistency and bugs fix libibnetdisc API consistency and bugs fix libibnetdisc API consistency and bugs fix libibnetdisc API consistency and bugs fix libibnetdisc API consistency and bugs split out ibnd_fabric_t fields that are only used during a scan remove 'dist' field from ibnd_node_t, which was virtually not used Dale Purdy (1): Add LMC support to DOR routing Doron Shoham (3): saquery: fix -c arguement osm_port.c: check if op_vls = 0 before max_op_vls comparison ibportstate: fixes for width option Eli Dorfman (Voltaire) (6): opensm/osm_lid_mgr.c bug in opensm LID assignment infiniband-diags: Do not change port state on SubnSet infiniband_diags/ibportstate.c perform reset in a single mad opensm/osm_sa.c add attribute and component mask to error message opensm/osm_sa_node_record.c use comp mask to match by LID or GUID Fix IB network discovery from switch node. Hal Rosenstock (127): opensm/PerfMgr: Change redir_tbl_size to num_ports for better clarity opensm: Add SuperMicro to list of recognized vendors infiniband-diags/saquery.c: Display attribute ID in hex rather than decimal opensm: Changes to spec and make files for updated release notes libibmad: Add support for SA PathRecord SL field opensm/PerfMgr: Remove some underbars from internal names infiniband-diags/ibnetdiscover.c: Cosmetic formatting changes opensm/PerfMgr DB: Remove leading underscores from internal names opensm/osm_perfmgr_db.c: Remove unneeded initialization in perfmgr_db_print_by_name opensm/PerfMgr: Cosmetic changes opensm/osm_port.c: Remove error number from debug level log message opensm/PerfMgr: Reduce host name length opensm/osm_port.c: Change log level of Invalid OP_VLS 0 message to VERBOSE infiniband-diags/ibdiag_common.c: Eliminate compile warning on x86_64 archs libibmad/resolve.c: Determine SL properly opensm/osm_mesh.c: Use define rather than hard coded constant opensm/osm_ucast_lash.c: Fix commentary typo opensm/osm_ucast_lash.c: Use calloc rather than malloc/memset libibmad/resolve.c: Remove compile warning on x86_64 arch opensm/osm_ucast_lash.(h c): Replace memory allocation by array opensm/osm_ucast_lash.c: Remove some unused code opensm/libvendor/osm_vendor_(ibumad mlx)_sa.c: Handle malloc failure in __osmv_send_sa_req opensm/man/osmtest.8: Update email address infiniband-diags/ibroute.c: Fix format and typo in printf opensm/osmtest: Fix some typos opensm/osm_ucast_mgr.c: Add error numbers to some error log messages opensm/osm_sa_(path multipath)_record.c: Fix typo in a couple of log messages opensm/osm_mcast_tbl.c: In osm_mcast_tbl_init, use calloc opensm/osm_switch.c: In osm_switch_prepare_path_rebuild, use realloc opensm/osm_ucast_ftree.c: Made error numbers unique in some log messages infiniband-diags/README: Update saquery dependencies opensm/osm_qos_policy.c: Use proper size in malloc in osm_qos_policy_vlarb_scope_create opensm/osm_qos_policy.c: Use calloc rather than malloc/memset infiniband-diags/ibtracert.c: Fix comparison to SwitchInfo.MulticastFDBCap in switch_mclookup infiniband-diags/ibroute.c: Fix typo in IBWARN message opensm/osm_mesh.c: Improve VL utilization infiniband-diags/ibtracert.c: Fix linearcap test in switch_lookup opensm/lash: Set minimum VL for LASH to use opensm/main.c: Display LASH start VL when specified opensm/osm_subnet.c: Format lash_start_vl consistent with other uint8 items opensm/osm_ucast_lash.c: Minor cleanups opensm: Add ability to configure SMSL opensm/main.c: Display SMSL when specified libibmad/mad.c: Validate DR hop count for SMPs and handle errors opensm/osm_ucast_lash.c: Handle malloc failures better opensm/osm_lash: Fix use after free problem in osm_mesh_node_delete opensm/ib_types.h: Update ib_mad_is_response description opensm: osm_dr_path_extend can fail due to invalid hop count opensm/osm_path.h: In osm_dr_path_init, only copy needed part of path opensm/osm_ucast_lash.c: Remove osm_mesh_node_delete call from switch_delete opensm/osm_path.h: Fix osm_dr_path_extend ret
[ewg] Re: [PATCH] IB/perftest: increase CQ size to avoid overflow
Ralph Campbell wrote: When running the ib_send_bw test bidirectional, there can be ctx->rx_depth + ctx->tx_depth/2 number of outstanding posted work queue entries but the (shared) completion queue is only size ctx->rx_depth. This can lead to completions being lost and the test hanging. Since memory isn't limited in the microbenchmark test and I think the ctx->tx_depth/2 send limit should probably be ctx->tx_depth, make the CQ size be ctx->rx_depth + ctx->tx_depth. Signed-off-by: Ralph Campbell diff --git a/send_bw.c b/send_bw.c index f842fb9..9624cfe 100644 --- a/send_bw.c +++ b/send_bw.c @@ -489,7 +489,8 @@ static struct pingpong_context *pp_init_ctx(struct ibv_device *ib_dev, } } - ctx->cq = ibv_create_cq(ctx->context, ctx->rx_depth, NULL, ctx->channel, 0); + ctx->cq = ibv_create_cq(ctx->context, ctx->rx_depth + ctx->tx_depth, + NULL, ctx->channel, 0); if (!ctx->cq) { fprintf(stderr, "Couldn't create CQ\n"); return NULL; Applied . Thanks Ralph. Regards, Ido ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
Re: [ewg] OMPI / OFED 1.5
Jeff Squyres wrote: Vlad -- I was remiss in making a ~jsquyres/ofed_1_5 directory on the OF server. I made one now and put the Open MPI 1.3.3 SRPM in it (with corresponding latest.txt). Updated in ofabuild, Regards, Vladimir ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
[ewg] ofa_1_5_kernel 20091101-0200 daily build status
This email was generated automatically, please do not reply git_url: git://git.openfabrics.org/ofed_1_5/linux-2.6.git git_branch: ofed_kernel_1_5 Common build parameters: Passed: Passed on i686 with linux-2.6.18 Passed on i686 with linux-2.6.19 Passed on i686 with linux-2.6.21.1 Passed on i686 with linux-2.6.26 Passed on i686 with linux-2.6.24 Passed on i686 with linux-2.6.22 Passed on i686 with linux-2.6.27 Passed on x86_64 with linux-2.6.16.60-0.54.5-smp Passed on x86_64 with linux-2.6.16.60-0.21-smp Passed on x86_64 with linux-2.6.18 Passed on x86_64 with linux-2.6.18-128.el5 Passed on x86_64 with linux-2.6.18-164.el5 Passed on x86_64 with linux-2.6.20 Passed on x86_64 with linux-2.6.19 Passed on x86_64 with linux-2.6.18-93.el5 Passed on x86_64 with linux-2.6.21.1 Passed on x86_64 with linux-2.6.24 Passed on x86_64 with linux-2.6.22 Passed on x86_64 with linux-2.6.26 Passed on x86_64 with linux-2.6.27 Passed on x86_64 with linux-2.6.25 Passed on x86_64 with linux-2.6.27.19-5-smp Passed on x86_64 with linux-2.6.9-67.ELsmp Passed on x86_64 with linux-2.6.9-78.ELsmp Passed on ia64 with linux-2.6.21.1 Passed on ia64 with linux-2.6.19 Passed on ia64 with linux-2.6.18 Passed on ia64 with linux-2.6.24 Passed on ia64 with linux-2.6.23 Passed on ia64 with linux-2.6.22 Passed on ia64 with linux-2.6.26 Passed on ia64 with linux-2.6.25 Passed on ppc64 with linux-2.6.18 Passed on ppc64 with linux-2.6.19 Failed: ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
[ewg] Re: [ANNOUNCE] uDAPL v2 - dapl-2.0.24 release
Davis, Arlin R wrote: Vlad, please pull new package into latest OFED 1.5 build and install the following: dapl-2.0.24-1 dapl-utils-2.0.24-1 dapl-devel-2.0.24-1 dapl-debuginfo-2.0.24-1 compat-dapl-1.2.14-1 compat-dapl-devel-1.2.14-1 See http://www.openfabrics.org/downloads/dapl/ more details. -arlin Done, Regards, Vladimir ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg