OFED 1.5.2 beta1 is available

OFED 1.5.2-beta1 is available

Notes:

The tarball is available on:
http://www.openfabrics.org/downloads/OFED/ofed-1.5.2/OFED-1.5.2-beta1.tgz

To get BUILD_ID run ofed_info

Please report any issues in bugzilla https://bugs.openfabrics.org/ for OFED 1.5.2

Vladimir & Tziporet

========================================================================
Supported Platforms and Operating Systems
---------------------------------------------
o CPU architectures:
- x86_64
- x86
- ppc64
- ia64

o Linux Operating Systems:
- RedHat EL4 up7 2.6.9-78.ELsmp
- RedHat EL4 up8 2.6.9-89.ELsmp
- RedHat EL5 up3 2.6.18-128.el5
- RedHat EL5 up4 2.6.18-164.el5
- RedHat EL5 up5 2.6.18-194.el5
- SLES10 SP2 2.6.16.60-0.21-smp
- SLES10 SP3 2.6.16.60-0.54-smp
- SLES11 2.6.27.19-5-default
- OEL 4 up7 2.6.9-78.ELsmp
- OEL 4 up8 2.6.9-89.ELsmp
- CentOS5.3 2.6.18-128.el5
- CentOS5.4 2.6.18-164.el5
- Fedora Core12 2.6.31.5-127.fc12 *
- OpenSuSE 11.2 2.6.31.5-0.1-default *
- kernel.org 2.6.29, 2.6.30,
2.6.31 and 2.6.32 *

* Minimal QA for these versions

Main changes from 1.5.1:
===========================
1. Updated packages:
- Management
Using latest daily builds from http://www.openfabrics.org/downloads/management/daily
- Updated libnes

     libnes-1.0.1-0.1.g89ea0ee.tar.gz
   - Updated libsdp
     libsdp-1.1.101-0.3.gc767eee.tar.gz
   - Updated perftest
     perftest-1.2.4-0.15.g82b7e29.tar.gz
   - Updated mpitests
     mpitests-3.2-923.src.rpm

2. Added RHEL5.5 support
3. Use files under /etc/modprobe.d/ instead of /etc/modprobe.conf
4. Bug fixes

commit 8986850c7198bf32fc3d87884c72875a272fdb5e

Author: Vladimir Sokolovsky <vlad-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Sun May 2 15:08:03 2010 +0300

mlx4_en.conf: load mlx4_en when mlx4_core is loaded

commit b03fe7f8623c2c4d9a716f90a848ab679d6b10a2

Author: Vladimir Sokolovsky <vlad-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Sun May 2 10:33:58 2010 +0300

ofa_kernel.spec: Stop using depecated /etc/modprobe.conf

commit ac488f9096cd4d146f5a99fba48c17f9d4235239

Author: Eli Cohen <eli-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Thu Apr 29 15:14:49 2010 +0300

mlx4: Fix layout of QPC.AH

commit 3aab0ebcfd1bb676b754cf9ca69a9b2609dd847e

Author: Eli Cohen <eli-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Thu Apr 29 14:36:39 2010 +0300

rdma_cm: Add default translation from ToS to SL

Add a default translation from for ToS to SL for rdma_cm over RoCE. The default translation is dictated by the value of the module paramter, def_prec2sl. Any value of ToS will be translated to this value. In the future we will add a more generic translation table.

commit 489dfc902a7b1e8f60ad996fec00113642301225

Author: Ralph Campbell (QLogic) <[email protected]>
Date:   Wed Apr 28 16:12:30 2010 -0700

IB/qib: more fixes for OFED 1.5.2

More serdes tuning changes for interop.
Fix a MR reference count bug with UC RDMA writes with immediate. Fix a deadlock race when registering more than one device with DCA.

commit 1270cdb8ed5f16079a2cdf83a73bc6e5df9c9333

Author: Vladimir Sokolovsky <vlad-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Wed Apr 28 16:38:07 2010 +0300

RDMA/ucma: Add option to manually set IB path

Export rdma_set_ib_paths to user space to allow applications to manually set the IB path used for connections. This allows alternative ways for a user space application or library to obtain path record information, including retrieving path information from cached data, avoiding direct interaction with the IB SA. The IB SA is a single, centralized entity that can limit scaling on large clusters running MPI applications.

Future changes to the rdma cm can expand on this framework to support the full range of features allowed by the IB CM, such as separate forward and reverse paths and APM.

commit e0ed697e60968670d4badfb1c8aa9125065d1823
Merge: e59e4bc 7b1e558
Author: Vladimir Sokolovsky <vlad-VPRAkNaXOzVS1MOuV/[email protected]>
Date: Tue Apr 27 10:03:13 2010 +0300

Merge branch 'ofed_kernel_1_5' of git://sofa.openfabrics.org/~ctung/ofed-1.5 into ofed_kernel_1_5

commit e59e4bc5fe76a5a8d9bdfa9407746d5227873bd9

Author: Ralph Campbell (QLogic) <[email protected]>
Date:   Mon Apr 26 10:58:37 2010 -0700

IB/qib: fix qib backport to RHEL 5.5

commit 30f0b496eefe781cb098ddfc774ff35b99217042

Author: Amir Vadai <amirv-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Mon Apr 26 14:53:23 2010 +0300

sdp: fix a leak when ib_post_xxx fail + small fixes

commit e3bf1a8d2da1759611f985493e79c61a0311e5d2

Author: Eldad Zinger <eldadz-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Mon Apr 26 10:48:48 2010 +0300

sdp: on device removal, ref count taken so that socket won't be destructed

On device removal, socket resources that are based on device are destroyed, but the socket itself will be destroyed later, upon user request.

commit 93bc358d9895e27474c35527188165a4d1188ca5
Merge: 8c94349 6c80f9c
Author: Vladimir Sokolovsky <vlad-VPRAkNaXOzVS1MOuV/[email protected]>
Date: Sun Apr 25 16:13:44 2010 +0300

Merge remote branch 'amirv/ofed_kernel_1_5' into ofed_kernel_1_5

commit 6c80f9ce0605f57c06629b2e4bec81752a78d0e1

Author: Amir Vadai <amirv-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Thu Apr 22 11:26:57 2010 +0300

sdp: use max number of SGE from HW capabilities

Instead of using hard coded max number of SGE's take it from device capabilities.

commit 4bc6920c83fb63a510fa86dec5de95ba6fc6dfa8

Author: Amir Vadai <amirv-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Sun Apr 25 11:55:29 2010 +0300

sdp: Fix a hang when ib_post_recv is failed

commit 0d56a06939d7c03091c26bd1deeb689a6cbba2f5

Author: Amir Vadai <amirv-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Sun Apr 25 11:50:44 2010 +0300

sdp: fix compilation warning on debug prints

commit 8c94349521f73054bfcb08748f33ad815defd854

Author: Eli Cohen <eli-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Sun Apr 25 10:41:37 2010 +0300

Bug fix in counters

Take the port number from qp->port rather than attr->port_num. qp->port is update when modifying from reset to init. For special QPs it is updated at creation.

commit b322c88f10bb2212101346e39523e32ce9f71739
Merge: 6d6ac62 6a54ece
Author: Vladimir Sokolovsky <vlad-VPRAkNaXOzVS1MOuV/[email protected]>
Date: Thu Apr 22 17:29:08 2010 +0300

Merge remote branch 'amirv/ofed_kernel_1_5' into ofed_kernel_1_5

commit 6d6ac621f1ddec038cd73685fc11c60dff76f706

Author: Vladimir Sokolovsky <vlad-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Thu Apr 22 16:27:38 2010 +0300

Update support for enhanced atomic operations

Patches updated with the version committed to the upstream kernel.

commit 6a54ece400371447a383c59597e961cb6d51977a

Author: Eldad Zinger <eldadz-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Thu Apr 22 10:34:59 2010 +0300

sdp: sdp_bzcopy_thresh module parameter removal

Field [bzcopy_thresh] inside 'struct sdp_sock' changed to [zcopy_thresh] and the ability to disable zcopy per-socket was added.

commit aaf5a11f6b1d466b5aaf2c9dd40b43dbe8f389bc

Author: Ralph Campbell (QLogic) <[email protected]>
Date:   Tue Apr 20 17:20:16 2010 -0700

IB/qib: backport fixes for OFED 1.5.2

Update the backport patches for OFED 1.5.2.

commit eceda2cf579365ee90fcb9c5c4ca647a16484777
Merge: b35c420 b48db55
Author: Vladimir Sokolovsky <vlad-VPRAkNaXOzVS1MOuV/[email protected]>
Date: Tue Apr 20 14:57:14 2010 +0300

    Merge branch 'ofed_kernel_1_5' of git://git.openfabrics.org/~ralphc/linux-2.6/ into ofed_kernel_1_5

commit b48db55de53486afcdf0422d79684627045731de

Author: Ralph Campbell (QLogic) <[email protected]>
Date:   Mon Apr 19 11:35:04 2010 -0700

IB/qib: fixes for OFED 1.5.2

This updates the QDR HCA serdes settings to improve reliability and is needed for the next IB interoperability event.

commit b35c4207dce433c1e3b623d9bdd4259f474e38e2

Author: Eldad Zinger <eldadz-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Sun Apr 18 16:56:03 2010 +0300

sdp: BUG2017 - better initialization implementation for ssk->nagle_timer

commit e79bd3e79785a6974feace5acea1e43fb1b7cd03

Author: Eldad Zinger <eldadz-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Sun Apr 18 14:34:33 2010 +0300

sdp: fix for brutal device removing

Kernel always crashed in the following test case:
user program: socket+bind+listen+accept. socket accpted.
shell: rmmod mlx4_ib
user program: close <<< CRASH
The fix closes any socket that its ib_device is the device being removed.

commit 5a9dbaf793268ecf4d44f53fdca42edda0f54e16

Author: Amir Vadai <amirv-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Thu Apr 15 11:57:11 2010 +0300

sdp: Don't try to allocate FMR larger than RLIMIT_MEMLOCK

During ZCopy. If don't have CAP_IPC_LOCK capability and current max number of locked pages is smaller than the buffer size, split the send into small fragments.

commit f988c01bf259206e9575bcc65d3d98e700f10911

Author: Amir Vadai <amirv-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Thu Apr 15 11:52:19 2010 +0300

sdp: Don't count sdp header twice when calculating size_goal

sizeof(struct sdp_bsdh) is included inside skb->len. Ignore it when calculating maximum payload of the skb.
This mistake caused every BCopy send of 32K (and its multiples) to be split and thus got bad performance.

commit e233ae1662b884deb9bdf77193e7a2b2a0f0b7d3

Author: Eldad Zinger <eldadz-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Tue Apr 13 13:39:41 2010 +0300

sdp: timeout for abortive close updated

SDP_FIN_WAIT_TIMEOUT updated from 10[sec] to 60[sec] to comply with TCP TCP_FIN_TIMEOUT.

commit 6cac8adf70072d141d403203693af9d94adb0e24

Author: Eli Cohen <eli-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Thu Apr 15 15:08:13 2010 +0300

Fix counters for RoCE

counter index is assigned in modify from init to rtr.

commit e5414cccaa13e6dd80d8d6fc3dafe95355facdef

Author: Amir Vadai <amirv-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Mon Apr 12 15:42:32 2010 +0300

sdp: module parameter to disable SDP over ROCEE

Since ROCEE added, SDP always try to connect, even if link layer of type Ethernet. This make 'both' mode in libsdp useless.

Added a module paramter to disable it by default

commit 9a6ef6793c83b1587984d919a84ed5c2ca19a0c0

Author: Eldad Zinger <eldadz-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Thu Apr 8 10:09:37 2010 +0300

sdp: added differentiation between bind failures of sdp.

When bind()ing in mode 'BOTH', bind(sdp_sock) might fail if:
1. the IP&port is already bounded.
2. the IP is not part of IB network.
previous implementation returned errno=EADDRINUSE either way.

Only the first case should fail the bind(), the second is legitimate because the TCP socket will hanle the connection.
This fix corresponds to a fix in libsdp.

commit feaa7f15f651c35978f8c2a3e08642d2530fce0b

Author: Eldad Zinger <eldadz-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Thu Mar 18 10:53:56 2010 +0200

sdp: BUG1727 - there is no point of using zcopy when credits are not available.

if credits are not available, data won't be transferred anyway, so it would be better to use bcopy.

commit 8f358b2307d4eaa816fc8bec3dda365399754530

Author: Eldad Zinger <eldadz-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Wed Mar 31 16:02:03 2010 +0300

sdp: BUG1992 - enable transmission of credits update when tx_credits == 1

According to spec, if one credit is available, we can only send messages that provide additional credits and also do not contain data payload.

commit 2ac23ff8ac947397ad6866cdf215809eb2511e93

Author: Eldad Zinger <eldadz-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Wed Mar 31 15:27:38 2010 +0300

sdp: unnecessary 'if' statement canceled.

commit 1c2405885e0e5595cdc531a84400458c4e0c70bb

Author: Eldad Zinger <eldadz-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Thu Mar 18 13:31:08 2010 +0200

sdp: BUG1727 - fixed select(2) behavior on a new nonblocking socket.

when calling for select(2) after socket(2) on a nonblocking socket, select(2) should return 'writable'.

commit f9f0b4d008fa53407d9a64a30c2789c6ce4cdf6e

Author: Eldad Zinger <eldadz-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Wed Mar 17 15:24:28 2010 +0200

sdp: BUG1727 - sdp_destroy_work() and sdp_connect() interfere with each other.

This fix was done in order to make sure that sdp_destroy_work() is done before sdp_connect() is active.

commit 88e95a2bf83596150919a077ea2e371f08812dbb

Author: Amir Vadai <amirv-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Sun Mar 14 16:36:32 2010 +0200

sdp: support iovlen > 1 in zcopy

commit 9cb9abb8501c96633d6ab144ccb1ff3880e12f74

Author: Amir Vadai <amirv-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Sun Mar 14 16:35:38 2010 +0200

sdp: make sdp_socket.h available to user applications

commit 7d39db269d341c8faf846b4ade0413c5dbb543f0

Author: Amir Vadai <amirv-VPRAkNaXOzVS1MOuV/[email protected]>
Date:   Thu Apr 1 10:28:37 2010 +0300

sdp: enable FMR pool cache

commit 7b1e5589a307f0783ff0b360a778bc6d32ab710a

Author: Chien Tung <[email protected]>
Date:   Thu Mar 25 10:33:30 2010 -0500

RDMA/nes: correct cap.max_inline_data assignment in nes_query_qp

commit 888aeb78804be9c0e89acb18622f49ff8cc40073

Author: Chien Tung <[email protected]>
Date:   Thu Mar 25 09:09:36 2010 -0500

RDMA/nes: disable dynamic interrupt moderation by default

commit 63079c7ac1161a090cdd9feef757118c9d56cc72
Merge: 17badd7 bf34b58
Author: Vladimir Sokolovsky <vlad-VPRAkNaXOzVS1MOuV/[email protected]>
Date: Thu Mar 25 11:01:55 2010 +0200

Merge remote branch 'vu/ofed_kernel_1_5' into ofed_kernel_1_5

commit bf34b58acf19fdbef467aa9bb70b37d840f3bdc7

Author: Vu Pham <vu-OnC3O1emZK9LiOotz6YBiwC/[email protected]>
Date:   Wed Mar 24 16:36:53 2010 -0700

srp: solving dead_lock issue during unload/reboot system

bug_id,"bug_severity","priority","op_sys","assigned_to","bug_status","resolution","short_short_desc"
1727,"normal","P3","SLES 11","[email protected]","RESOLVED","FIXED","Select fails after connect to invalid address"
1847,"normal","P3","All","[email protected]","RESOLVED","FIXED","zcopy send with iovlen > 1 is not supported"
1872,"normal","P3","All","[email protected]","RESOLVED","FIXED","failed to send packets less than 4080 byte by zcopy" 1909,"normal","P3","All","[email protected]","RESOLVED","FIXED","Enable a trivial way to include sdp_socket.h"
1965,"major","P3","RHEL 5","[email protected]","CLOSED","FIXED","Bonding mlx4_en: ping does not resume after failover between 10G ports" 1984,"normal","P3","All","[email protected]","RESOLVED","FIXED","Bug in accept()"
1988,"major","P2","All","[email protected]","RESOLVED","FIXED","[OFED-1.5.1-rc4] - Softirq seen on one of the nodes after ~48 hours while running 2 instances each of IMB-MPI1+ OSU+Presta test together on 64 bit platforms" 2027,"normal","P3","RHEL 5","[email protected]","CLOSED","FIXED","SDP not respecting # SGEs as reported from HW"
2041,"normal","P3","Other","[email protected]","RESOLVED","FIXED","QIB compilation fails on RHEL5.5"

I would like to add support for IB ACM path resolution (i.e. user space path record caching) into the 1.5.2 release. Vlad pulled in the necessary kernel patch from 2.6.33 a few days ago. All other changes are in user space -- an updated librdmacm release and a release of ibacm -- and must be enabled to be used.

I hope to complete testing this week.

- Sean

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html