Release 2.0.32

Signed-off-by: Arlin Davis <arlin.r.davis@intel.com>

diff --git a/ChangeLog b/ChangeLog
index 51992d37f81f38b4c9eb1b184cba610238c8b3e4..e5eda7ab610b4d582cc3975bc7d75a298193e0fd 100644 (file)
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,114 @@
+commit ef80301e5799bd5657071e6bf4a1787711b5c878
+Author: Arlin Davis <arlin.r.davis@intel.com>
+Date:   Sat Feb 12 12:03:19 2011 -0800
+
+    cma: reduce output log level in disconnect from WARN to CM_WARN
+    
+    Signed-off-by: Arlin Davis <arlin.r.davis@intel.com>
+
+commit 0ca585d4c3e5043e3000786eee7293c3fc4818e3
+Author: Arlin Davis <arlin.r.davis@intel.com>
+Date:   Sat Feb 12 11:46:08 2011 -0800
+
+    ucm: delay freeing of active side UD cm object in case RTU is dropped
+    
+    The ucm was freeing the UD CM object to quickly so a retried REPLY
+    was dropped and the passive side never received the AH info via RTU.
+    Keep active side UD cm objects on work queue until QP is destroyed
+    so RTU can be resent if necessary.
+    
+    Signed-off-by: Arlin Davis <arlin.r.davis@intel.com>
+
+commit a4de5f009dfe303788e08e26d46c5e0c68f20e6e
+Author: Arlin Davis <arlin.r.davis@intel.com>
+Date:   Sat Feb 12 11:36:35 2011 -0800
+
+    ucm: cm object needs to be on work queue before req sent on wire
+    
+    With this delay in cm object queuing there is potential for replies
+    being dropped coming back with a NO MATCH. Start with INIT
+    state and queue it up, move to state REP_PENDING when
+    sending out on the wire to start request timer.
+    
+    Signed-off-by: Arlin Davis <arlin.r.davis@intel.com>
+
+commit 342d0eb23c7da570d6f787474389377f706e8f14
+Author: Arlin Davis <arlin.r.davis@intel.com>
+Date:   Fri Feb 4 14:41:45 2011 -0800
+
+    ucm, scm: remove use of usec_sleep delays and use events for disc and destroy
+    
+    use pthread mutex when processing and waiting for disconnect completions
+    and for CM object destruction. Add f_event, d_event to cm object.
+    
+    Signed-off-by: Arlin Davis <arlin.r.davis@intel.com>
+
+commit 02a298eca46a00e3cd26f694d54b8194efc231d1
+Author: Arlin Davis <arlin.r.davis@intel.com>
+Date:   Thu Jan 20 18:31:30 2011 -0800
+
+    common: reduce default max inline data size because of performance anomaly
+    
+    Increasing max inline causes small message rates to decrease from
+    4M/sec when set to 64 to about 1M/sec when set to 400. This has
+    been observed on latest mlx4 adapters. Set default to 64 until resolved.
+    
+    Signed-off-by: Arlin Davis <arlin.r.davis@intel.com>
+
+commit 14402cefc4ad90f4bd826f60bdeef7ad82909041
+Author: Arlin Davis <arlin.r.davis@intel.com>
+Date:   Thu Jan 20 11:05:02 2011 -0800
+
+    common: dapls_evd_dto_wait() dbg message should print status and not errno
+    
+    Signed-off-by: Arlin Davis <arlin.r.davis@intel.com>
+
+commit 7050cd7b4756ed98072513ee9c3672f0cb63a97d
+Author: Arlin Davis <arlin.r.davis@intel.com>
+Date:   Mon Jan 17 14:43:58 2011 -0800
+
+    ucm, scm: exchange max_qp_rd_atom and limit outstanding requests
+    
+    exchange and add proper checking to limit outstanding
+    rdma reads and atomics. Use one of the reserve bytes
+    in CM message protocol to exchange limits and reset
+    EP attribute rdma_out and set QP RTS attribute properly.
+    
+    Signed-off-by: Arlin Davis <arlin.r.davis@intel.com>
+
+commit 080b172cf26ab3202cfb4a1f231e849ea911473f
+Author: Arlin Davis <arlin.r.davis@intel.com>
+Date:   Tue Jan 4 20:22:12 2011 -0800
+
+    scm: retry socket connect on ECONNREFUSED under heavy load
+    
+    with large scale workloads a linux server starts rejecting
+    socket connect requests. Add retry logic for connection refused
+    errors.
+    
+    increasing net.ipv4.tcp_max_syn_backlog to 2048 will also reduce the
+    chance of these errors when scaling up.
+    
+    Signed-off-by: Arlin Davis <arlin.r.davis@intel.com>
+
+commit fee5b15e67c60d9d39005ff99655f2c41c418d37
+Author: Arlin Davis <arlin.r.davis@intel.com>
+Date:   Tue Jan 4 16:25:29 2011 -0800
+
+    common: qp modify RTR using wrong ep attribute parameter for dest_rd_atomic
+    
+    max_rdma_read_in should be used instead of max_rdma_read_out
+    
+    Signed-off-by: Arlin Davis <arlin.r.davis@intel.com>
+
+commit b2df3373fec12f16df712e6e2e35d8a3799f3452
+Author: Arlin Davis <arlin.r.davis@intel.com>
+Date:   Fri Dec 10 14:26:07 2010 -0800
+
+    Release 2.0.31
+    
+    Signed-off-by: Arlin Davis <arlin.r.davis@intel.com>
+
 commit 2bebd3f72ecd73e2df80ca4712f0f470647f196b
 Author: Arlin Davis <arlin.r.davis@intel.com>
 Date:   Fri Dec 10 14:19:45 2010 -0800

diff --git a/configure.in b/configure.in
index 1b2af468b431e607501f3db8c05f978642427db1..c78942f5df2d90bb9eeea2d66140cefcc1ebc7ec 100644 (file)
--- a/configure.in
+++ b/configure.in
@@ -1,11 +1,11 @@
 dnl Process this file with autoconf to produce a configure script.
 
 AC_PREREQ(2.57)
-AC_INIT(dapl, 2.0.31, linux-rdma@vger.kernel.org)
+AC_INIT(dapl, 2.0.32, linux-rdma@vger.kernel.org)
 AC_CONFIG_SRCDIR([dat/udat/udat.c])
 AC_CONFIG_AUX_DIR(config)
 AM_CONFIG_HEADER(config.h)
-AM_INIT_AUTOMAKE(dapl, 2.0.31)
+AM_INIT_AUTOMAKE(dapl, 2.0.32)
 
 AM_PROG_LIBTOOL
 

diff --git a/dapl.spec.in b/dapl.spec.in
index d824f24d5bdb0349d12b19a44544685d1f9f67a7..38cfa69c483f8a274891efe5b400b1135cbde4ef 100644 (file)
--- a/dapl.spec.in
+++ b/dapl.spec.in
@@ -144,6 +144,9 @@ fi
 %{_mandir}/man5/*.5*
 
 %changelog
+* Sun Feb 13 2011 Arlin Davis <ardavis@ichips.intel.com> - 2.0.32
+- DAT/DAPL Version 2.0.32 Release 1, OFED 1.5.3 GA 
+
 * Fri Dec 10 2010 Arlin Davis <ardavis@ichips.intel.com> - 2.0.31
 - DAT/DAPL Version 2.0.31 Release 1, OFED 1.5.3